{ "cells": [ { "cell_type": "markdown", "id": "be3c1872", "metadata": {}, "source": [ "# AI-based OCR Benchmark Notebook\n", "\n", "This notebook benchmarks **AI-based OCR models** on scanned PDF documents/images in Spanish.\n", "It excludes traditional OCR engines like Tesseract that require external installations." ] }, { "cell_type": "code", "execution_count": 1, "id": "6a1e98fe", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Requirement already satisfied: pip in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (25.3)\n", "Note: you may need to restart the kernel to use updated packages.\n", "Requirement already satisfied: jupyter in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (1.1.1)\n", "Requirement already satisfied: notebook in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jupyter) (7.5.0)\n", "Requirement already satisfied: jupyter-console in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jupyter) (6.6.3)\n", "Requirement already satisfied: nbconvert in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jupyter) (7.16.6)\n", "Requirement already satisfied: ipykernel in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jupyter) (7.1.0)\n", "Requirement already satisfied: ipywidgets in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jupyter) (8.1.8)\n", "Requirement already satisfied: jupyterlab in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jupyter) (4.5.0)\n", "Requirement already satisfied: comm>=0.1.1 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipykernel->jupyter) (0.2.3)\n", "Requirement already satisfied: debugpy>=1.6.5 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipykernel->jupyter) (1.8.17)\n", "Requirement already satisfied: ipython>=7.23.1 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipykernel->jupyter) (9.8.0)\n", "Requirement already satisfied: jupyter-client>=8.0.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipykernel->jupyter) (8.6.3)\n", "Requirement already satisfied: jupyter-core!=5.0.*,>=4.12 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipykernel->jupyter) (5.9.1)\n", "Requirement already satisfied: matplotlib-inline>=0.1 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipykernel->jupyter) (0.2.1)\n", "Requirement already satisfied: nest-asyncio>=1.4 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipykernel->jupyter) (1.6.0)\n", "Requirement already satisfied: packaging>=22 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipykernel->jupyter) (25.0)\n", "Requirement already satisfied: psutil>=5.7 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipykernel->jupyter) (7.1.3)\n", "Requirement already satisfied: pyzmq>=25 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipykernel->jupyter) (27.1.0)\n", "Requirement already satisfied: tornado>=6.2 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipykernel->jupyter) (6.5.2)\n", "Requirement already satisfied: traitlets>=5.4.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipykernel->jupyter) (5.14.3)\n", "Requirement already satisfied: colorama>=0.4.4 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipython>=7.23.1->ipykernel->jupyter) (0.4.6)\n", "Requirement already satisfied: decorator>=4.3.2 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipython>=7.23.1->ipykernel->jupyter) (5.2.1)\n", "Requirement already satisfied: ipython-pygments-lexers>=1.0.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipython>=7.23.1->ipykernel->jupyter) (1.1.1)\n", "Requirement already satisfied: jedi>=0.18.1 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipython>=7.23.1->ipykernel->jupyter) (0.19.2)\n", "Requirement already satisfied: prompt_toolkit<3.1.0,>=3.0.41 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipython>=7.23.1->ipykernel->jupyter) (3.0.52)\n", "Requirement already satisfied: pygments>=2.11.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipython>=7.23.1->ipykernel->jupyter) (2.19.2)\n", "Requirement already satisfied: stack_data>=0.6.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipython>=7.23.1->ipykernel->jupyter) (0.6.3)\n", "Requirement already satisfied: typing_extensions>=4.6 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipython>=7.23.1->ipykernel->jupyter) (4.15.0)\n", "Requirement already satisfied: wcwidth in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from prompt_toolkit<3.1.0,>=3.0.41->ipython>=7.23.1->ipykernel->jupyter) (0.2.14)\n", "Requirement already satisfied: parso<0.9.0,>=0.8.4 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jedi>=0.18.1->ipython>=7.23.1->ipykernel->jupyter) (0.8.5)\n", "Requirement already satisfied: python-dateutil>=2.8.2 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jupyter-client>=8.0.0->ipykernel->jupyter) (2.9.0.post0)\n", "Requirement already satisfied: platformdirs>=2.5 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jupyter-core!=5.0.*,>=4.12->ipykernel->jupyter) (4.5.1)\n", "Requirement already satisfied: six>=1.5 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from python-dateutil>=2.8.2->jupyter-client>=8.0.0->ipykernel->jupyter) (1.17.0)\n", "Requirement already satisfied: executing>=1.2.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from stack_data>=0.6.0->ipython>=7.23.1->ipykernel->jupyter) (2.2.1)\n", "Requirement already satisfied: asttokens>=2.1.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from stack_data>=0.6.0->ipython>=7.23.1->ipykernel->jupyter) (3.0.1)\n", "Requirement already satisfied: pure-eval in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from stack_data>=0.6.0->ipython>=7.23.1->ipykernel->jupyter) (0.2.3)\n", "Requirement already satisfied: widgetsnbextension~=4.0.14 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipywidgets->jupyter) (4.0.15)\n", "Requirement already satisfied: jupyterlab_widgets~=3.0.15 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipywidgets->jupyter) (3.0.16)\n", "Requirement already satisfied: async-lru>=1.0.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jupyterlab->jupyter) (2.0.5)\n", "Requirement already satisfied: httpx<1,>=0.25.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jupyterlab->jupyter) (0.28.1)\n", "Requirement already satisfied: jinja2>=3.0.3 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jupyterlab->jupyter) (3.1.6)\n", "Requirement already satisfied: jupyter-lsp>=2.0.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jupyterlab->jupyter) (2.3.0)\n", "Requirement already satisfied: jupyter-server<3,>=2.4.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jupyterlab->jupyter) (2.17.0)\n", "Requirement already satisfied: jupyterlab-server<3,>=2.28.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jupyterlab->jupyter) (2.28.0)\n", "Requirement already satisfied: notebook-shim>=0.2 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jupyterlab->jupyter) (0.2.4)\n", "Requirement already satisfied: setuptools>=41.1.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jupyterlab->jupyter) (65.5.0)\n", "Requirement already satisfied: anyio in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from httpx<1,>=0.25.0->jupyterlab->jupyter) (4.12.0)\n", "Requirement already satisfied: certifi in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from httpx<1,>=0.25.0->jupyterlab->jupyter) (2025.11.12)\n", "Requirement already satisfied: httpcore==1.* in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from httpx<1,>=0.25.0->jupyterlab->jupyter) (1.0.9)\n", "Requirement already satisfied: idna in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from httpx<1,>=0.25.0->jupyterlab->jupyter) (3.11)\n", "Requirement already satisfied: h11>=0.16 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from httpcore==1.*->httpx<1,>=0.25.0->jupyterlab->jupyter) (0.16.0)\n", "Requirement already satisfied: argon2-cffi>=21.1 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jupyter-server<3,>=2.4.0->jupyterlab->jupyter) (25.1.0)\n", "Requirement already satisfied: jupyter-events>=0.11.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jupyter-server<3,>=2.4.0->jupyterlab->jupyter) (0.12.0)\n", "Requirement already satisfied: jupyter-server-terminals>=0.4.4 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jupyter-server<3,>=2.4.0->jupyterlab->jupyter) (0.5.3)\n", "Requirement already satisfied: nbformat>=5.3.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jupyter-server<3,>=2.4.0->jupyterlab->jupyter) (5.10.4)\n", "Requirement already satisfied: overrides>=5.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jupyter-server<3,>=2.4.0->jupyterlab->jupyter) (7.7.0)\n", "Requirement already satisfied: prometheus-client>=0.9 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jupyter-server<3,>=2.4.0->jupyterlab->jupyter) (0.23.1)\n", "Requirement already satisfied: pywinpty>=2.0.1 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jupyter-server<3,>=2.4.0->jupyterlab->jupyter) (3.0.2)\n", "Requirement already satisfied: send2trash>=1.8.2 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jupyter-server<3,>=2.4.0->jupyterlab->jupyter) (1.8.3)\n", "Requirement already satisfied: terminado>=0.8.3 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jupyter-server<3,>=2.4.0->jupyterlab->jupyter) (0.18.1)\n", "Requirement already satisfied: websocket-client>=1.7 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jupyter-server<3,>=2.4.0->jupyterlab->jupyter) (1.9.0)\n", "Requirement already satisfied: babel>=2.10 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jupyterlab-server<3,>=2.28.0->jupyterlab->jupyter) (2.17.0)\n", "Requirement already satisfied: json5>=0.9.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jupyterlab-server<3,>=2.28.0->jupyterlab->jupyter) (0.12.1)\n", "Requirement already satisfied: jsonschema>=4.18.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jupyterlab-server<3,>=2.28.0->jupyterlab->jupyter) (4.25.1)\n", "Requirement already satisfied: requests>=2.31 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jupyterlab-server<3,>=2.28.0->jupyterlab->jupyter) (2.32.5)\n", "Requirement already satisfied: argon2-cffi-bindings in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from argon2-cffi>=21.1->jupyter-server<3,>=2.4.0->jupyterlab->jupyter) (25.1.0)\n", "Requirement already satisfied: MarkupSafe>=2.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jinja2>=3.0.3->jupyterlab->jupyter) (3.0.3)\n", "Requirement already satisfied: attrs>=22.2.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jsonschema>=4.18.0->jupyterlab-server<3,>=2.28.0->jupyterlab->jupyter) (25.4.0)\n", "Requirement already satisfied: jsonschema-specifications>=2023.03.6 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jsonschema>=4.18.0->jupyterlab-server<3,>=2.28.0->jupyterlab->jupyter) (2025.9.1)\n", "Requirement already satisfied: referencing>=0.28.4 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jsonschema>=4.18.0->jupyterlab-server<3,>=2.28.0->jupyterlab->jupyter) (0.37.0)\n", "Requirement already satisfied: rpds-py>=0.7.1 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jsonschema>=4.18.0->jupyterlab-server<3,>=2.28.0->jupyterlab->jupyter) (0.30.0)\n", "Requirement already satisfied: python-json-logger>=2.0.4 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jupyter-events>=0.11.0->jupyter-server<3,>=2.4.0->jupyterlab->jupyter) (4.0.0)\n", "Requirement already satisfied: pyyaml>=5.3 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jupyter-events>=0.11.0->jupyter-server<3,>=2.4.0->jupyterlab->jupyter) (6.0.2)\n", "Requirement already satisfied: rfc3339-validator in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jupyter-events>=0.11.0->jupyter-server<3,>=2.4.0->jupyterlab->jupyter) (0.1.4)\n", "Requirement already satisfied: rfc3986-validator>=0.1.1 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jupyter-events>=0.11.0->jupyter-server<3,>=2.4.0->jupyterlab->jupyter) (0.1.1)\n", "Requirement already satisfied: fqdn in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jsonschema[format-nongpl]>=4.18.0->jupyter-events>=0.11.0->jupyter-server<3,>=2.4.0->jupyterlab->jupyter) (1.5.1)\n", "Requirement already satisfied: isoduration in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jsonschema[format-nongpl]>=4.18.0->jupyter-events>=0.11.0->jupyter-server<3,>=2.4.0->jupyterlab->jupyter) (20.11.0)\n", "Requirement already satisfied: jsonpointer>1.13 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jsonschema[format-nongpl]>=4.18.0->jupyter-events>=0.11.0->jupyter-server<3,>=2.4.0->jupyterlab->jupyter) (3.0.0)\n", "Requirement already satisfied: rfc3987-syntax>=1.1.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jsonschema[format-nongpl]>=4.18.0->jupyter-events>=0.11.0->jupyter-server<3,>=2.4.0->jupyterlab->jupyter) (1.1.0)\n", "Requirement already satisfied: uri-template in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jsonschema[format-nongpl]>=4.18.0->jupyter-events>=0.11.0->jupyter-server<3,>=2.4.0->jupyterlab->jupyter) (1.3.0)\n", "Requirement already satisfied: webcolors>=24.6.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jsonschema[format-nongpl]>=4.18.0->jupyter-events>=0.11.0->jupyter-server<3,>=2.4.0->jupyterlab->jupyter) (25.10.0)\n", "Requirement already satisfied: beautifulsoup4 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from nbconvert->jupyter) (4.14.3)\n", "Requirement already satisfied: bleach!=5.0.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from bleach[css]!=5.0.0->nbconvert->jupyter) (6.3.0)\n", "Requirement already satisfied: defusedxml in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from nbconvert->jupyter) (0.7.1)\n", "Requirement already satisfied: jupyterlab-pygments in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from nbconvert->jupyter) (0.3.0)\n", "Requirement already satisfied: mistune<4,>=2.0.3 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from nbconvert->jupyter) (3.1.4)\n", "Requirement already satisfied: nbclient>=0.5.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from nbconvert->jupyter) (0.10.2)\n", "Requirement already satisfied: pandocfilters>=1.4.1 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from nbconvert->jupyter) (1.5.1)\n", "Requirement already satisfied: webencodings in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from bleach!=5.0.0->bleach[css]!=5.0.0->nbconvert->jupyter) (0.5.1)\n", "Requirement already satisfied: tinycss2<1.5,>=1.1.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from bleach[css]!=5.0.0->nbconvert->jupyter) (1.4.0)\n", "Requirement already satisfied: fastjsonschema>=2.15 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from nbformat>=5.3.0->jupyter-server<3,>=2.4.0->jupyterlab->jupyter) (2.21.2)\n", "Requirement already satisfied: charset_normalizer<4,>=2 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from requests>=2.31->jupyterlab-server<3,>=2.28.0->jupyterlab->jupyter) (3.4.4)\n", "Requirement already satisfied: urllib3<3,>=1.21.1 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from requests>=2.31->jupyterlab-server<3,>=2.28.0->jupyterlab->jupyter) (2.6.0)\n", "Requirement already satisfied: lark>=1.2.2 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from rfc3987-syntax>=1.1.0->jsonschema[format-nongpl]>=4.18.0->jupyter-events>=0.11.0->jupyter-server<3,>=2.4.0->jupyterlab->jupyter) (1.3.1)\n", "Requirement already satisfied: cffi>=1.0.1 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from argon2-cffi-bindings->argon2-cffi>=21.1->jupyter-server<3,>=2.4.0->jupyterlab->jupyter) (2.0.0)\n", "Requirement already satisfied: pycparser in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from cffi>=1.0.1->argon2-cffi-bindings->argon2-cffi>=21.1->jupyter-server<3,>=2.4.0->jupyterlab->jupyter) (2.23)\n", "Requirement already satisfied: soupsieve>=1.6.1 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from beautifulsoup4->nbconvert->jupyter) (2.8)\n", "Requirement already satisfied: arrow>=0.15.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from isoduration->jsonschema[format-nongpl]>=4.18.0->jupyter-events>=0.11.0->jupyter-server<3,>=2.4.0->jupyterlab->jupyter) (1.4.0)\n", "Requirement already satisfied: tzdata in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from arrow>=0.15.0->isoduration->jsonschema[format-nongpl]>=4.18.0->jupyter-events>=0.11.0->jupyter-server<3,>=2.4.0->jupyterlab->jupyter) (2025.2)\n", "Note: you may need to restart the kernel to use updated packages.\n", "Requirement already satisfied: ipywidgets in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (8.1.8)\n", "Requirement already satisfied: comm>=0.1.3 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipywidgets) (0.2.3)\n", "Requirement already satisfied: ipython>=6.1.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipywidgets) (9.8.0)\n", "Requirement already satisfied: traitlets>=4.3.1 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipywidgets) (5.14.3)\n", "Requirement already satisfied: widgetsnbextension~=4.0.14 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipywidgets) (4.0.15)\n", "Requirement already satisfied: jupyterlab_widgets~=3.0.15 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipywidgets) (3.0.16)\n", "Requirement already satisfied: colorama>=0.4.4 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipython>=6.1.0->ipywidgets) (0.4.6)\n", "Requirement already satisfied: decorator>=4.3.2 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipython>=6.1.0->ipywidgets) (5.2.1)\n", "Requirement already satisfied: ipython-pygments-lexers>=1.0.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipython>=6.1.0->ipywidgets) (1.1.1)\n", "Requirement already satisfied: jedi>=0.18.1 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipython>=6.1.0->ipywidgets) (0.19.2)\n", "Requirement already satisfied: matplotlib-inline>=0.1.5 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipython>=6.1.0->ipywidgets) (0.2.1)\n", "Requirement already satisfied: prompt_toolkit<3.1.0,>=3.0.41 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipython>=6.1.0->ipywidgets) (3.0.52)\n", "Requirement already satisfied: pygments>=2.11.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipython>=6.1.0->ipywidgets) (2.19.2)\n", "Requirement already satisfied: stack_data>=0.6.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipython>=6.1.0->ipywidgets) (0.6.3)\n", "Requirement already satisfied: typing_extensions>=4.6 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipython>=6.1.0->ipywidgets) (4.15.0)\n", "Requirement already satisfied: wcwidth in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from prompt_toolkit<3.1.0,>=3.0.41->ipython>=6.1.0->ipywidgets) (0.2.14)\n", "Requirement already satisfied: parso<0.9.0,>=0.8.4 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jedi>=0.18.1->ipython>=6.1.0->ipywidgets) (0.8.5)\n", "Requirement already satisfied: executing>=1.2.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from stack_data>=0.6.0->ipython>=6.1.0->ipywidgets) (2.2.1)\n", "Requirement already satisfied: asttokens>=2.1.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from stack_data>=0.6.0->ipython>=6.1.0->ipywidgets) (3.0.1)\n", "Requirement already satisfied: pure-eval in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from stack_data>=0.6.0->ipython>=6.1.0->ipywidgets) (0.2.3)\n", "Note: you may need to restart the kernel to use updated packages.\n", "Requirement already satisfied: ipykernel in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (7.1.0)\n", "Requirement already satisfied: comm>=0.1.1 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipykernel) (0.2.3)\n", "Requirement already satisfied: debugpy>=1.6.5 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipykernel) (1.8.17)\n", "Requirement already satisfied: ipython>=7.23.1 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipykernel) (9.8.0)\n", "Requirement already satisfied: jupyter-client>=8.0.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipykernel) (8.6.3)\n", "Requirement already satisfied: jupyter-core!=5.0.*,>=4.12 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipykernel) (5.9.1)\n", "Requirement already satisfied: matplotlib-inline>=0.1 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipykernel) (0.2.1)\n", "Requirement already satisfied: nest-asyncio>=1.4 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipykernel) (1.6.0)\n", "Requirement already satisfied: packaging>=22 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipykernel) (25.0)\n", "Requirement already satisfied: psutil>=5.7 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipykernel) (7.1.3)\n", "Requirement already satisfied: pyzmq>=25 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipykernel) (27.1.0)\n", "Requirement already satisfied: tornado>=6.2 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipykernel) (6.5.2)\n", "Requirement already satisfied: traitlets>=5.4.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipykernel) (5.14.3)\n", "Requirement already satisfied: colorama>=0.4.4 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipython>=7.23.1->ipykernel) (0.4.6)\n", "Requirement already satisfied: decorator>=4.3.2 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipython>=7.23.1->ipykernel) (5.2.1)\n", "Requirement already satisfied: ipython-pygments-lexers>=1.0.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipython>=7.23.1->ipykernel) (1.1.1)\n", "Requirement already satisfied: jedi>=0.18.1 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipython>=7.23.1->ipykernel) (0.19.2)\n", "Requirement already satisfied: prompt_toolkit<3.1.0,>=3.0.41 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipython>=7.23.1->ipykernel) (3.0.52)\n", "Requirement already satisfied: pygments>=2.11.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipython>=7.23.1->ipykernel) (2.19.2)\n", "Requirement already satisfied: stack_data>=0.6.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipython>=7.23.1->ipykernel) (0.6.3)\n", "Requirement already satisfied: typing_extensions>=4.6 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ipython>=7.23.1->ipykernel) (4.15.0)\n", "Requirement already satisfied: wcwidth in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from prompt_toolkit<3.1.0,>=3.0.41->ipython>=7.23.1->ipykernel) (0.2.14)\n", "Requirement already satisfied: parso<0.9.0,>=0.8.4 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jedi>=0.18.1->ipython>=7.23.1->ipykernel) (0.8.5)\n", "Requirement already satisfied: python-dateutil>=2.8.2 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jupyter-client>=8.0.0->ipykernel) (2.9.0.post0)\n", "Requirement already satisfied: platformdirs>=2.5 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jupyter-core!=5.0.*,>=4.12->ipykernel) (4.5.1)\n", "Requirement already satisfied: six>=1.5 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from python-dateutil>=2.8.2->jupyter-client>=8.0.0->ipykernel) (1.17.0)\n", "Requirement already satisfied: executing>=1.2.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from stack_data>=0.6.0->ipython>=7.23.1->ipykernel) (2.2.1)\n", "Requirement already satisfied: asttokens>=2.1.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from stack_data>=0.6.0->ipython>=7.23.1->ipykernel) (3.0.1)\n", "Requirement already satisfied: pure-eval in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from stack_data>=0.6.0->ipython>=7.23.1->ipykernel) (0.2.3)\n", "Note: you may need to restart the kernel to use updated packages.\n" ] } ], "source": [ "%pip install --upgrade pip\n", "%pip install --upgrade jupyter\n", "%pip install --upgrade ipywidgets\n", "%pip install --upgrade ipykernel" ] }, { "cell_type": "code", "execution_count": 2, "id": "13103c58", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Requirement already satisfied: transformers in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (4.57.3)\n", "Requirement already satisfied: pillow in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (12.0.0)\n", "Requirement already satisfied: paddleocr in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (3.3.2)\n", "Requirement already satisfied: hf_xet in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (1.2.0)\n", "Requirement already satisfied: paddlepaddle in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (3.2.2)\n", "Requirement already satisfied: jiwer in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (4.0.0)\n", "Requirement already satisfied: rich in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (14.2.0)\n", "Requirement already satisfied: filelock in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from transformers) (3.20.0)\n", "Requirement already satisfied: huggingface-hub<1.0,>=0.34.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from transformers) (0.36.0)\n", "Requirement already satisfied: numpy>=1.17 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from transformers) (2.3.5)\n", "Requirement already satisfied: packaging>=20.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from transformers) (25.0)\n", "Requirement already satisfied: pyyaml>=5.1 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from transformers) (6.0.2)\n", "Requirement already satisfied: regex!=2019.12.17 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from transformers) (2025.11.3)\n", "Requirement already satisfied: requests in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from transformers) (2.32.5)\n", "Requirement already satisfied: tokenizers<=0.23.0,>=0.22.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from transformers) (0.22.1)\n", "Requirement already satisfied: safetensors>=0.4.3 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from transformers) (0.7.0)\n", "Requirement already satisfied: tqdm>=4.27 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from transformers) (4.67.1)\n", "Requirement already satisfied: fsspec>=2023.5.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from huggingface-hub<1.0,>=0.34.0->transformers) (2025.12.0)\n", "Requirement already satisfied: typing-extensions>=3.7.4.3 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from huggingface-hub<1.0,>=0.34.0->transformers) (4.15.0)\n", "Requirement already satisfied: paddlex<3.4.0,>=3.3.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from paddlex[ocr-core]<3.4.0,>=3.3.0->paddleocr) (3.3.10)\n", "Requirement already satisfied: aistudio-sdk>=0.3.5 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from paddlex<3.4.0,>=3.3.0->paddlex[ocr-core]<3.4.0,>=3.3.0->paddleocr) (0.3.8)\n", "Requirement already satisfied: chardet in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from paddlex<3.4.0,>=3.3.0->paddlex[ocr-core]<3.4.0,>=3.3.0->paddleocr) (5.2.0)\n", "Requirement already satisfied: colorlog in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from paddlex<3.4.0,>=3.3.0->paddlex[ocr-core]<3.4.0,>=3.3.0->paddleocr) (6.10.1)\n", "Requirement already satisfied: modelscope>=1.28.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from paddlex<3.4.0,>=3.3.0->paddlex[ocr-core]<3.4.0,>=3.3.0->paddleocr) (1.32.0)\n", "Requirement already satisfied: pandas>=1.3 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from paddlex<3.4.0,>=3.3.0->paddlex[ocr-core]<3.4.0,>=3.3.0->paddleocr) (2.3.3)\n", "Requirement already satisfied: prettytable in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from paddlex<3.4.0,>=3.3.0->paddlex[ocr-core]<3.4.0,>=3.3.0->paddleocr) (3.17.0)\n", "Requirement already satisfied: py-cpuinfo in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from paddlex<3.4.0,>=3.3.0->paddlex[ocr-core]<3.4.0,>=3.3.0->paddleocr) (9.0.0)\n", "Requirement already satisfied: pydantic>=2 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from paddlex<3.4.0,>=3.3.0->paddlex[ocr-core]<3.4.0,>=3.3.0->paddleocr) (2.12.5)\n", "Requirement already satisfied: ruamel.yaml in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from paddlex<3.4.0,>=3.3.0->paddlex[ocr-core]<3.4.0,>=3.3.0->paddleocr) (0.18.16)\n", "Requirement already satisfied: ujson in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from paddlex<3.4.0,>=3.3.0->paddlex[ocr-core]<3.4.0,>=3.3.0->paddleocr) (5.11.0)\n", "Requirement already satisfied: imagesize in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from paddlex[ocr-core]<3.4.0,>=3.3.0->paddleocr) (1.4.1)\n", "Requirement already satisfied: opencv-contrib-python==4.10.0.84 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from paddlex[ocr-core]<3.4.0,>=3.3.0->paddleocr) (4.10.0.84)\n", "Requirement already satisfied: pyclipper in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from paddlex[ocr-core]<3.4.0,>=3.3.0->paddleocr) (1.4.0)\n", "Requirement already satisfied: pypdfium2>=4 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from paddlex[ocr-core]<3.4.0,>=3.3.0->paddleocr) (5.1.0)\n", "Requirement already satisfied: python-bidi in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from paddlex[ocr-core]<3.4.0,>=3.3.0->paddleocr) (0.6.7)\n", "Requirement already satisfied: shapely in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from paddlex[ocr-core]<3.4.0,>=3.3.0->paddleocr) (2.1.2)\n", "Requirement already satisfied: httpx in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from paddlepaddle) (0.28.1)\n", "Requirement already satisfied: protobuf>=3.20.2 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from paddlepaddle) (6.33.2)\n", "Requirement already satisfied: opt-einsum==3.3.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from paddlepaddle) (3.3.0)\n", "Requirement already satisfied: networkx in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from paddlepaddle) (3.6)\n", "Requirement already satisfied: click>=8.1.8 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jiwer) (8.2.1)\n", "Requirement already satisfied: rapidfuzz>=3.9.7 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jiwer) (3.14.3)\n", "Requirement already satisfied: markdown-it-py>=2.2.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from rich) (4.0.0)\n", "Requirement already satisfied: pygments<3.0.0,>=2.13.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from rich) (2.19.2)\n", "Requirement already satisfied: psutil in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from aistudio-sdk>=0.3.5->paddlex<3.4.0,>=3.3.0->paddlex[ocr-core]<3.4.0,>=3.3.0->paddleocr) (7.1.3)\n", "Requirement already satisfied: bce-python-sdk in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from aistudio-sdk>=0.3.5->paddlex<3.4.0,>=3.3.0->paddlex[ocr-core]<3.4.0,>=3.3.0->paddleocr) (0.9.55)\n", "Requirement already satisfied: colorama in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from click>=8.1.8->jiwer) (0.4.6)\n", "Requirement already satisfied: mdurl~=0.1 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from markdown-it-py>=2.2.0->rich) (0.1.2)\n", "Requirement already satisfied: setuptools in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from modelscope>=1.28.0->paddlex<3.4.0,>=3.3.0->paddlex[ocr-core]<3.4.0,>=3.3.0->paddleocr) (65.5.0)\n", "Requirement already satisfied: urllib3>=1.26 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from modelscope>=1.28.0->paddlex<3.4.0,>=3.3.0->paddlex[ocr-core]<3.4.0,>=3.3.0->paddleocr) (2.6.0)\n", "Requirement already satisfied: python-dateutil>=2.8.2 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from pandas>=1.3->paddlex<3.4.0,>=3.3.0->paddlex[ocr-core]<3.4.0,>=3.3.0->paddleocr) (2.9.0.post0)\n", "Requirement already satisfied: pytz>=2020.1 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from pandas>=1.3->paddlex<3.4.0,>=3.3.0->paddlex[ocr-core]<3.4.0,>=3.3.0->paddleocr) (2025.2)\n", "Requirement already satisfied: tzdata>=2022.7 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from pandas>=1.3->paddlex<3.4.0,>=3.3.0->paddlex[ocr-core]<3.4.0,>=3.3.0->paddleocr) (2025.2)\n", "Requirement already satisfied: annotated-types>=0.6.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from pydantic>=2->paddlex<3.4.0,>=3.3.0->paddlex[ocr-core]<3.4.0,>=3.3.0->paddleocr) (0.7.0)\n", "Requirement already satisfied: pydantic-core==2.41.5 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from pydantic>=2->paddlex<3.4.0,>=3.3.0->paddlex[ocr-core]<3.4.0,>=3.3.0->paddleocr) (2.41.5)\n", "Requirement already satisfied: typing-inspection>=0.4.2 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from pydantic>=2->paddlex<3.4.0,>=3.3.0->paddlex[ocr-core]<3.4.0,>=3.3.0->paddleocr) (0.4.2)\n", "Requirement already satisfied: six>=1.5 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from python-dateutil>=2.8.2->pandas>=1.3->paddlex<3.4.0,>=3.3.0->paddlex[ocr-core]<3.4.0,>=3.3.0->paddleocr) (1.17.0)\n", "Requirement already satisfied: charset_normalizer<4,>=2 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from requests->transformers) (3.4.4)\n", "Requirement already satisfied: idna<4,>=2.5 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from requests->transformers) (3.11)\n", "Requirement already satisfied: certifi>=2017.4.17 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from requests->transformers) (2025.11.12)\n", "Requirement already satisfied: pycryptodome>=3.8.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from bce-python-sdk->aistudio-sdk>=0.3.5->paddlex<3.4.0,>=3.3.0->paddlex[ocr-core]<3.4.0,>=3.3.0->paddleocr) (3.23.0)\n", "Requirement already satisfied: future>=0.6.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from bce-python-sdk->aistudio-sdk>=0.3.5->paddlex<3.4.0,>=3.3.0->paddlex[ocr-core]<3.4.0,>=3.3.0->paddleocr) (1.0.0)\n", "Requirement already satisfied: anyio in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from httpx->paddlepaddle) (4.12.0)\n", "Requirement already satisfied: httpcore==1.* in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from httpx->paddlepaddle) (1.0.9)\n", "Requirement already satisfied: h11>=0.16 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from httpcore==1.*->httpx->paddlepaddle) (0.16.0)\n", "Requirement already satisfied: wcwidth in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from prettytable->paddlex<3.4.0,>=3.3.0->paddlex[ocr-core]<3.4.0,>=3.3.0->paddleocr) (0.2.14)\n", "Requirement already satisfied: ruamel.yaml.clib>=0.2.7 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ruamel.yaml->paddlex<3.4.0,>=3.3.0->paddlex[ocr-core]<3.4.0,>=3.3.0->paddleocr) (0.2.15)\n", "Note: you may need to restart the kernel to use updated packages.\n", "Requirement already satisfied: pandas in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (2.3.3)\n", "Requirement already satisfied: numpy>=1.23.2 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from pandas) (2.3.5)\n", "Requirement already satisfied: python-dateutil>=2.8.2 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from pandas) (2.9.0.post0)\n", "Requirement already satisfied: pytz>=2020.1 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from pandas) (2025.2)\n", "Requirement already satisfied: tzdata>=2022.7 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from pandas) (2025.2)\n", "Requirement already satisfied: six>=1.5 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from python-dateutil>=2.8.2->pandas) (1.17.0)\n", "Note: you may need to restart the kernel to use updated packages.\n", "Requirement already satisfied: matplotlib in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (3.10.7)\n", "Requirement already satisfied: contourpy>=1.0.1 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from matplotlib) (1.3.3)\n", "Requirement already satisfied: cycler>=0.10 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from matplotlib) (0.12.1)\n", "Requirement already satisfied: fonttools>=4.22.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from matplotlib) (4.61.0)\n", "Requirement already satisfied: kiwisolver>=1.3.1 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from matplotlib) (1.4.9)\n", "Requirement already satisfied: numpy>=1.23 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from matplotlib) (2.3.5)\n", "Requirement already satisfied: packaging>=20.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from matplotlib) (25.0)\n", "Requirement already satisfied: pillow>=8 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from matplotlib) (12.0.0)\n", "Requirement already satisfied: pyparsing>=3 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from matplotlib) (3.2.5)\n", "Requirement already satisfied: python-dateutil>=2.7 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from matplotlib) (2.9.0.post0)\n", "Requirement already satisfied: six>=1.5 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from python-dateutil>=2.7->matplotlib) (1.17.0)\n", "Note: you may need to restart the kernel to use updated packages.\n", "Requirement already satisfied: seaborn in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (0.13.2)\n", "Requirement already satisfied: numpy!=1.24.0,>=1.20 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from seaborn) (2.3.5)\n", "Requirement already satisfied: pandas>=1.2 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from seaborn) (2.3.3)\n", "Requirement already satisfied: matplotlib!=3.6.1,>=3.4 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from seaborn) (3.10.7)\n", "Requirement already satisfied: contourpy>=1.0.1 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (1.3.3)\n", "Requirement already satisfied: cycler>=0.10 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (0.12.1)\n", "Requirement already satisfied: fonttools>=4.22.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (4.61.0)\n", "Requirement already satisfied: kiwisolver>=1.3.1 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (1.4.9)\n", "Requirement already satisfied: packaging>=20.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (25.0)\n", "Requirement already satisfied: pillow>=8 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (12.0.0)\n", "Requirement already satisfied: pyparsing>=3 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (3.2.5)\n", "Requirement already satisfied: python-dateutil>=2.7 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (2.9.0.post0)\n", "Requirement already satisfied: pytz>=2020.1 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from pandas>=1.2->seaborn) (2025.2)\n", "Requirement already satisfied: tzdata>=2022.7 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from pandas>=1.2->seaborn) (2025.2)\n", "Requirement already satisfied: six>=1.5 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from python-dateutil>=2.7->matplotlib!=3.6.1,>=3.4->seaborn) (1.17.0)\n", "Note: you may need to restart the kernel to use updated packages.\n" ] } ], "source": [ "# Install necessary packages\n", "%pip install transformers pillow paddleocr hf_xet paddlepaddle jiwer rich\n", "\n", "\n", "\n", "# Data analysis and visualization\n", "%pip install pandas\n", "%pip install matplotlib\n", "%pip install seaborn" ] }, { "cell_type": "code", "execution_count": 1, "id": "ae33632a", "metadata": {}, "outputs": [], "source": [ "# Imports\n", "import os, json\n", "import numpy as np\n", "import pandas as pd\n", "import matplotlib.pyplot as plt\n", "\n", "import re\n", "from datetime import datetime\n", "\n", "from rich.console import Console\n", "import colorama\n", "\n", "colorama.just_fix_windows_console()\n", "# Tell Ray Tune to use a Jupyter-compatible console\n", "console = Console(force_jupyter=True)" ] }, { "cell_type": "markdown", "id": "0e00f1b0", "metadata": {}, "source": [ "## 1 Configuration" ] }, { "cell_type": "code", "execution_count": 2, "id": "8bfa3329", "metadata": {}, "outputs": [], "source": [ "PDF_FOLDER = './dataset' # Folder containing PDF files\n", "OUTPUT_FOLDER = 'results'\n", "os.makedirs(OUTPUT_FOLDER, exist_ok=True)" ] }, { "cell_type": "code", "execution_count": 3, "id": "8bd4ca23", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "c:\\Users\\Sergio\\Desktop\\MastersThesis\\src\\dataset\n", "c:\\Users\\Sergio\\Desktop\\MastersThesis\\src\\paddle_ocr_tuning.py\n", "c:\\Users\\Sergio\\Desktop\\MastersThesis\\src\n" ] } ], "source": [ "PDF_FOLDER_ABS = os.path.abspath(PDF_FOLDER) # ./instructions -> C:\\...\\instructions\n", "SCRIPT_ABS = os.path.abspath(\"paddle_ocr_tuning.py\") # paddle_ocr_tuning.py -> C:\\...\\paddle_ocr_tuning.py\n", "SCRIPT_DIR = os.path.dirname(SCRIPT_ABS)\n", "\n", "print(PDF_FOLDER_ABS)\n", "print(SCRIPT_ABS)\n", "print(SCRIPT_DIR)" ] }, { "cell_type": "code", "execution_count": 4, "id": "9c658b58", "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "c:\\Users\\Sergio\\Desktop\\MastersThesis\\.venv\\Lib\\site-packages\\paddle\\utils\\cpp_extension\\extension_utils.py:718: UserWarning: No ccache found. Please be aware that recompiling all source files may be required. You can download and install ccache from: https://github.com/ccache/ccache/blob/master/doc/INSTALL.md\n", " warnings.warn(warning_message)\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Paddle version: 3.2.2\n", "GPU available: False\n", "GPU count: 0\n", "Current device: cpu\n" ] } ], "source": [ "import paddle\n", "\n", "print(\"Paddle version:\", paddle.__version__)\n", "print(\"GPU available:\", paddle.device.is_compiled_with_cuda())\n", "print(\"GPU count:\", paddle.device.cuda.device_count())\n", "print(\"Current device:\", paddle.device.get_device())" ] }, { "cell_type": "code", "execution_count": 7, "id": "243849b9", "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "\u001b[32mCreating model: ('PP-LCNet_x1_0_doc_ori', None)\u001b[0m\n", "\u001b[32mModel files already exist. Using cached files. To redownload, please delete the directory manually: `C:\\Users\\Sergio\\.paddlex\\official_models\\PP-LCNet_x1_0_doc_ori`.\u001b[0m\n", "\u001b[32mCreating model: ('UVDoc', None)\u001b[0m\n", "\u001b[32mModel files already exist. Using cached files. To redownload, please delete the directory manually: `C:\\Users\\Sergio\\.paddlex\\official_models\\UVDoc`.\u001b[0m\n", "\u001b[32mCreating model: ('PP-LCNet_x1_0_textline_ori', None)\u001b[0m\n", "\u001b[32mModel files already exist. Using cached files. To redownload, please delete the directory manually: `C:\\Users\\Sergio\\.paddlex\\official_models\\PP-LCNet_x1_0_textline_ori`.\u001b[0m\n", "\u001b[32mCreating model: ('PP-OCRv5_server_det', None)\u001b[0m\n", "\u001b[32mModel files already exist. Using cached files. To redownload, please delete the directory manually: `C:\\Users\\Sergio\\.paddlex\\official_models\\PP-OCRv5_server_det`.\u001b[0m\n", "\u001b[32mCreating model: ('PP-OCRv5_server_rec', None)\u001b[0m\n", "\u001b[32mModel files already exist. Using cached files. To redownload, please delete the directory manually: `C:\\Users\\Sergio\\.paddlex\\official_models\\PP-OCRv5_server_rec`.\u001b[0m\n" ] } ], "source": [ "# 3. PaddleOCR \n", "# https://www.paddleocr.ai/v3.0.0/en/version3.x/pipeline_usage/OCR.html?utm_source=chatgpt.com#21-command-line\n", "from paddleocr import PaddleOCR\n", "\n", "# Initialize with better settings for Spanish/Latin text\n", "# https://www.paddleocr.ai/main/en/version3.x/algorithm/PP-OCRv5/PP-OCRv5_multi_languages.html?utm_source=chatgpt.com#5-models-and-their-supported-languages\n", "paddleocr_model = PaddleOCR(\n", " text_detection_model_name=\"PP-OCRv5_server_det\",\n", " text_recognition_model_name=\"PP-OCRv5_server_rec\"\n", ")" ] }, { "cell_type": "code", "execution_count": 8, "id": "329da34a", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "3.3.2\n" ] } ], "source": [ "import paddleocr\n", "\n", "print(paddleocr.__version__)" ] }, { "cell_type": "code", "execution_count": 9, "id": "b1541bb6", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "c:\\Users\\Sergio\\Desktop\\MastersThesis\\.venv\\Lib\\site-packages\\paddleocr\n" ] } ], "source": [ "# 1) Locate the installed PaddleOCR package\n", "pkg_dir = os.path.dirname(paddleocr.__file__)\n", "print(pkg_dir)" ] }, { "cell_type": "markdown", "id": "84c999e2", "metadata": {}, "source": [ "## 2 Helper Functions" ] }, { "cell_type": "code", "execution_count": 10, "id": "9596c7df", "metadata": {}, "outputs": [], "source": [ "from typing import List, Optional\n", "from paddle_ocr_tuning import evaluate_text, assemble_from_paddle_result\n", "from dataset_manager import ImageTextDataset" ] }, { "cell_type": "code", "execution_count": 11, "id": "b7c1bbf8", "metadata": {}, "outputs": [], "source": [ "from PIL import Image\n", "\n", "def show_page(img: Image.Image, scale: float = 1):\n", " \"\"\"\n", " Displays a smaller version of the image with text as a footer.\n", " \"\"\"\n", " # Compute plot size based on image dimensions (but without resizing the image)\n", " w, h = img.size\n", " figsize = (w * scale / 100, h * scale / 100) # convert pixels to inches approx\n", "\n", " fig, ax = plt.subplots(figsize=figsize)\n", " ax.imshow(img)\n", " ax.axis(\"off\")\n", "\n", "\n", " # Add OCR text below the image (footer)\n", " # plt.figtext(0.5, 0.02, text.strip(), wrap=True, ha='center', va='bottom', fontsize=10)\n", " plt.tight_layout()\n", " plt.show()" ] }, { "cell_type": "code", "execution_count": 12, "id": "b9d3fe25", "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "Índice\n", "1. Indicaciones generales 3\n", "1.1. Línea de discurso 3\n", "1.2. Estructura general y extensión del TFE 4\n", "1.3. Formatos y plantilla de trabajo 5\n", "1.4. Estética y estilo de redacción 7\n", "1.5. Normativa de citas 8\n", "2. Estructura del documento 9\n", "2.1. Resumen 10\n", "2.2. Organización del trabajo en grupo 11\n", "2.3. Introducción 11\n", "2.4. Contexto y estado del arte 13\n", "2.5. Objetivos concretos y metodología de trabajo 14\n", "2.6. Desarrollo específico de la contribución 17\n", "2.7. Conclusiones y trabajo futuro 20\n", "2.8. Referencias bibliográficas 21\n", "© Universidad Internacional de La Rioja (UNIR)\n", "2.8.1. Herramientas para buscar bibliografía 22\n", "2.9. Anexos 23\n", "2.10. Índice de acrónimos 24\n" ] } ], "source": [ "#test\n", "dataset = ImageTextDataset(PDF_FOLDER_ABS)\n", "img, txt = dataset[1]\n", "show_page(img, 0.15)\n", "print(txt)" ] }, { "cell_type": "code", "execution_count": 13, "id": "dcd27755", "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "Superior e inferior: 2,5 cm.\n", "Formato de párrafo en texto principal (estilo de la plantilla “Normal”):\n", " Calibri 12, justificado, interlineado 1,5, espacio entre párrafos 6 puntos\n", "anterior y 6 puntos posterior, sin sangría.\n", "Títulos:\n", " Primer nivel (estilo de la plantilla “Título 1”): Calibri Light 18, azul, justificado,\n", "interlineado 1,5, espacio entre párrafos 6 puntos anterior y 6 puntos\n", "posterior, sin sangría.\n", " Segundo nivel (estilo de la plantilla “Título 2”): Calibri Light 14, azul,\n", "justificado, interlineado 1,5, espacio entre párrafos 6 puntos anterior y 6\n", "puntos posterior, sin sangría.\n", " Tercer nivel (estilo de la plantilla “Título 3”: Calibri Light 12, justificado,\n", "interlineado 1,5, espacio entre párrafos 6 puntos anterior y 6 puntos\n", "posterior, sin sangría.\n", "Notas al pie:\n", " Calibri 10, justificado, interlineado sencillo, espacio entre párrafos 0 puntos\n", "anterior y 0 puntos posterior, sin sangría.\n", "Tablas y figuras:\n", " Título en la parte superior de la tabla o figura.\n", " Numeración tabla o figura (Tabla 1/ Figura1): Calibri 12, negrita, justificado.\n", " Nombre tabla o figura: Calibri 12, cursiva, justificado.\n", " Cuerpo: la tipografía de las tablas o figuras se pueden reducir hasta los 9\n", "puntos si estas contienen mucha información. Si la tabla o figura es muy\n", "grande, también se puede colocar en apaisado dentro de la hoja.\n", " Fuente de la tabla o figura en la parte inferior. Calibri 9,5, centrado.\n", "Encabezado y pie de página:\n", " Todas las páginas llevarán un encabezado con el nombre completo del\n", "estudiante y el título del TFE.\n", "© Universidad Internacional de La Rioja (UNIR)\n", " Todas las páginas llevarán también un pie de página con el número de página.\n", "Instrucciones para la redacción y elaboración del TFE\n", "6\n", "Máster Universitario en Inteligencia Artificial\n" ] } ], "source": [ "dataset = ImageTextDataset(PDF_FOLDER_ABS)\n", "img, txt = dataset[5]\n", "show_page(img, 0.15)\n", "print(txt)" ] }, { "cell_type": "markdown", "id": "e42cae29", "metadata": {}, "source": [ "## Run AI OCR Benchmark" ] }, { "cell_type": "code", "execution_count": 14, "id": "9b55c154", "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "ref: \n", "Superior e inferior: 2,5 cm.\n", "Formato de párrafo en texto principal (estilo de la plantilla “Normal”):\n", " Calibri 12, justificado, interlineado 1,5, espacio entre párrafos 6 puntos\n", "anterior y 6 puntos posterior, sin sangría.\n", "Títulos:\n", " Primer nivel (estilo de la plantilla “Título 1”): Calibri Light 18, azul, justificado,\n", "interlineado 1,5, espacio entre párrafos 6 puntos anterior y 6 puntos\n", "posterior, sin sangría.\n", " Segundo nivel (estilo de la plantilla “Título 2”): Calibri Light 14, azul,\n", "justificado, interlineado 1,5, espacio entre párrafos 6 puntos anterior y 6\n", "puntos posterior, sin sangría.\n", " Tercer nivel (estilo de la plantilla “Título 3”: Calibri Light 12, justificado,\n", "interlineado 1,5, espacio entre párrafos 6 puntos anterior y 6 puntos\n", "posterior, sin sangría.\n", "Notas al pie:\n", " Calibri 10, justificado, interlineado sencillo, espacio entre párrafos 0 puntos\n", "anterior y 0 puntos posterior, sin sangría.\n", "Tablas y figuras:\n", " Título en la parte superior de la tabla o figura.\n", " Numeración tabla o figura (Tabla 1/ Figura1): Calibri 12, negrita, justificado.\n", " Nombre tabla o figura: Calibri 12, cursiva, justificado.\n", " Cuerpo: la tipografía de las tablas o figuras se pueden reducir hasta los 9\n", "puntos si estas contienen mucha información. Si la tabla o figura es muy\n", "grande, también se puede colocar en apaisado dentro de la hoja.\n", " Fuente de la tabla o figura en la parte inferior. Calibri 9,5, centrado.\n", "Encabezado y pie de página:\n", " Todas las páginas llevarán un encabezado con el nombre completo del\n", "estudiante y el título del TFE.\n", "© Universidad Internacional de La Rioja (UNIR)\n", " Todas las páginas llevarán también un pie de página con el número de página.\n", "Instrucciones para la redacción y elaboración del TFE\n", "6\n", "Máster Universitario en Inteligencia Artificial\n", "paddle_text: \n", "Superior e inferior: 2,5 cm.\n", "Formato de párrafo en texto principal (estilo de la plantilla “Normal\"):\n", "Calibri 12, justificado, interlineado 1,5, espacio entre párrafos 6 puntos\n", "anterior y 6 puntos posterior, sin sangría.\n", "Títulos:\n", "Primer nivel (estilo de la plantillaTítulo 1\"): Calibri Light 18, azul, justificado,\n", "interlineado 1,5,espacio entre párrafos 6 puntos anterior y 6 puntos\n", "posterior, sin sangría.\n", "Segundo nivel (estilo de la plantilla Titulo 2\"): Calibri Light 14, azul,\n", "justificado, interlineado 1,5, espacio entre párrafos 6 puntos anterior y 6\n", "puntos posterior, sin sangría.\n", "Tercer nivel (estilo de la plantilla Título 3\": Calibri Light 12, justificado,\n", "interlineado 1,5,espacio entre párrafos 6 puntos anterior y 6 puntos\n", "posterior, sin sangría.\n", "Notas al pie:\n", "Calibri 10, justificado, interlineado sencillo, espacio entre párrafos O puntos\n", "anterior y O puntos posterior, sin sangra.\n", "Tablas y figuras:\n", "Título en la parte superior de la tabla o figura.\n", "Numeración tabla o figura (Tabla 1/ Figura1): Calibri 12, negrita, justificado.\n", "Nombre tabla o figura: Calibri 12, cursiva, justificado.\n", "Cuerpo: la tipografía de las tablas o figuras se pueden reducir hasta los 9\n", "puntos si estas contienen mucha información. Si la tabla o figura es muy\n", "grande, también se puede colocar en apaisado dentro de la hoja.\n", "Fuente de la tabla o figura en la parte inferior. Calibri 9,5, centrado.\n", "Encabezado y pie de página:\n", "Todas las páginas llevarán un encabezado con el nombre completo del\n", "estudiante y el título del TFE.\n", "© Universidad Internacional de La Rioja (UNiR)\n", "Todas las páginas llevarán también un pie de página con el número de página.\n", "Instrucciones para la redacción y elaboración del TFE\n", "Máster Universitario en Inteligencia Artificial 9\n" ] }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "ref: \n", "Los borradores intermedios deberán entregarse en formato Word. El documento final\n", "deberá depositarse en formato PDF.\n", "1.4. Estética y estilo de redacción\n", "Es fundamental que el TFE presente un aspecto elegante y correcto. Se trata de un\n", "trabajo académico y debe reflejar la madurez y el nivel formativo de una persona que\n", "ha finalizado un estudio de grado o postgrado. Ten en cuenta las siguientes\n", "recomendaciones en todas y cada una de las entregas que realices y, en especial, en\n", "el depósito final del documento:\n", " Verifica la originalidad del documento, asegurándote de que citas todas las\n", "fuentes consultadas y no existen textos de autoría ajena sin referenciar\n", "correctamente.\n", " Cuida la presentación del trabajo. Comprueba que formatos como tipo y tamaño\n", "de letra, número de páginas, encabezados, justificación de párrafos, interlineado,\n", "etc., son correctos.\n", " Revisa la ortografía y la redacción. Utiliza el corrector de Word para asegurarte de\n", "que no has dejado ninguna errata. Una lectura detenida del documento también\n", "te ayudará a detectar erratas, omisiones o redundancias. Si es posible, pide a\n", "alguien cercano que lo lea y te dé su opinión sobre la redacción. Presta especial\n", "atención a los siguientes aspectos:\n", "- Que los párrafos sigan un orden o hilo argumental lógico.\n", "- Que la información se presente de una manera que facilite su\n", "comprensión, definiendo los conceptos necesarios e incluyendo las citas\n", "bibliográficas pertinentes.\n", "- Elimina párrafos demasiado cortos. Cada párrafo debería tener al menos\n", "© Universidad Internacional de La Rioja (UNIR)\n", "tres oraciones.\n", "- Elimina frases superfluas y repeticiones de ideas.\n", "Instrucciones para la redacción y elaboración del TFE\n", "7\n", "Máster Universitario en Inteligencia Artificial\n", "paddle_text: \n", "Los borradores intermedios deberán entregarse en formato Word. El documento final\n", "deberá depositarse en formato PDf.\n", "1.4. Estética y estilo de redacción\n", "Es fundamental que el TFE presente un aspecto elegante y correcto. Se trata de un\n", "trabajo académico y debe reflejar la madurez y el nivel formativo de una persona que\n", "ha finalizado un estudio de grado o postgrado. Ten en cuenta las siguientes\n", "recomendaciones en todas y cada una de las entregas que realices y, en especial, en\n", "el deposito final del documento:\n", "Verifica la originalidad del documento,asegurándote de que citas todas las\n", "fuentes consultadas y no existen textos de autoría ajena sin referenciar\n", "correctamente.\n", "Cuida la presentación del trabajo. Comprueba que formatos como tipo y tamaño\n", "de letra, número de páginas, encabezados, justificación de párrafos, interlineado,\n", "etc., son correctos.\n", "Revisa la ortografía y la redacción. Utiliza el corrector de Word para asegurarte de\n", "que no has dejado ninguna errata. Una lectura detenida del documento también\n", "te ayudará a detectar erratas, omisiones o redundancias. Si es posible, pide a\n", "alguien cercano que lo lea y te dé su opinión sobre la redacción. Presta especial\n", "atención a los siguientes aspectos:\n", "Que los párrafos sigan un orden o hilo argumental lógico.\n", "Que la información se presente de una manera que facilite su\n", "comprensión, definiendo los conceptos necesarios e incluyendo las citas\n", "bibliograficas pertinentes.\n", "Elimina párrafos demasiado cortos. Cada párrafo debería tener al menos\n", "© Universidad Internacional de La Rioja (UNiR) tres oraciones.\n", "Elimina frases superfluas y repeticiones de ideas.\n", "Instrucciones para la redacción y elaboración del TfE 7\n", "Máster Universitario en Inteligencia Artificial\n" ] }, { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAWoAAAH3CAYAAACSIBV+AAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjcsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvTLEjVAAAAAlwSFlzAAAPYQAAD2EBqD+naQAAf+VJREFUeJztvQeUVdeV570rZ6qgoKDISYBIIgkJUM7BSlayLQepg3u67e52zzfzrZn5Znp6utdMu9vttt0O07Ys56AMkkARkZNAiJxTkUNRUDmHb/324xSPMqGQSuii9/+tBe/VDSfdc/9nn/Pu3jepra2tzYQQQkSW5E+6AEIIIc6PhFoIISKOhFoIISKOhFoIISKOhFoIISKOhFoIISKOhFoIISKOhFoIISKOhFoIISKOhFoIISKOhFoIISKOhFoIISKOhFoIISKOhFoIISKOhFoIISKOhFoIISKOhFoIISKOhFoIISKOhFoIISKOhFoIISKOhFoIISKOhFoIISKOhFoIISKOhFoIISKOhFoIISKOhFoIISKOhFoIISKOhFoIISKOhFoIISKOhFoIISKOhFoIISKOhFoIISKOhFoIISKOhFoIISKOhFoIISKOhFoIISKOhFoIISKOhFoIISKOhFoIISKOhFoIISKOhFoIISKOhFoIISKOhFoIISKOhFoIISKOhFoIISKOhFoIISKOhFoIISKOhFoIISKOhFoIISKOhFoIISKOhPoS0tbWZidPnrTDhw9beflJ//t8VFVVWXV1tW3bttVaW1s7nUdpaekF0/6oUI9Dhw5aFGlqarItW7Z0qs2OHTtmy5Yt9XPOBels3779rOnRzmVlZX9wPts7cw0aGxtt165dnTqWPMiLYy+mjh8F+hL/Pgr19fXt7dexnmy7mP6dqEioLyF0xp///Oe2cuVK27OnxP9GiPlsaWnx73xy89bW1trRo0ettPSYvfvuPN/HzUlHZ39NTU17p+ezubm5/fwdO3Z4mg0NDVZbGzsupMlxbOMz5BXSIv26ujrfx83F8fF5cDyDB/sPHjxoGzZsbK9bc3OTn8M/8uUzpBnqyPfYMXXtZSY90g1twbbTacbKyD++hzo0NTWeUa7T22PtQ9n4Oykp6Yx9oW78zXfyfPPNNyw1Nc2Skszz5/iO7UoZ5859x8tB+Tk3XnD27Nnj20Nbku66dets06ZNfj7tEfKLz4PtCO/ChQv8OPbH1zlcT8pOuuRRUlLieR46dMhqaqr/oI4dr1do+1CO+Gsa+mRo9/jrwTEcu3nzZtu5c6enzb54QeWYcM1DOuQR30dJiwGdz9mzZ7eXM/QH8pw3b561tsbatmP5+F5zqu4ft/ERZVI/6QJczsR3HG6YzkCHHThwgA0cONDefPNN75gzZsywFStW+M08evRoW7hwofXp09v69etveXm5dvToEXvllVe88z/00EP26quvWnJysl111VU2fvx4T/e111718kyfPsP27t1rV1xxhc2ZM9vLNWnSZFuwYIHl5eVZWdlxKyzs6d+HDx9uixYttJSUVC8DNyQDw7XXTrO1a9d4ejfeeJMNHTrU80B8OKalpdnGjh17Rju88847duDAQbe0i4uLrby83L74xS/anDlzLCUl2evCjX74MAJTa1/4whe8/tQjJSXFhgwZYjt2bLerr55qI0eO9HQ/+OADy87Obh9Q+DsrK8u/P/nkk5aamur7fvrTp61bt3yrq6u1e++911avXm1Hjhy2AQMG2K9//SsbOHAQpbTS0uM2Y8Z0t2CZ1dx//wO2detW69Gjh82dO9eOHz/uovDEE1+03NxcT/u3v/2tl+XIkSM+ANCOiMrtt9/h6cPBgwesoKDAZs6cad27F9iQIUNt584dVldXb2lpafbyyy/ZNddc6+mRDgL7wAMP2qxZsyw1NcWP45pzzfbv32+9evXyclVXV9kVV4ywTZs2WmZmll/rAwcOWE5Oti1ZstQHR+rNIJKbm+fi96Uvfcnbk7woD9u41t26dbP169e7OD766KPWt29fv27U58CB/TZs2DBrbW2z/fv3WX5+gQ0dOsQWLFhox4+X2s0332KLFy/y60vfpU/Qr2gPrm9RUS8bPXqMbdy4wduYPvqb3/zaevbs6eWjfblW7777rrcTs8kZM66zVatW+nncRnv37rN5897179OmTWvvX1zLDRs2+L1BuvStREQW9UcAQXzxxRfc8u0sbW2tVlsbs+wyMjLcAuGmQsA+97nP2ZVXXuk3wWc/+7ALAjdxz5697LHHHrPGxgZbs2aNixACwA0WIC0EjO2IJdNibhpuEG5+eOSRR/zvhx9+2MvMIDF27Di75557/GZApD7zmft8OaCystJFNH55IycnxzIzM23Xrt3W0nLmVJVzb7nlZuvXr59df/31LlhMdxHf9PR0LwMW1y233GqDBg308lFv6szUOtSVeoQBEMs/WLCIMMJE2YN1GKAs1A2h4ViEjIECazUtLd1v8MbGJhfxwYOHWHZ2jlVUVHgaI0eO8oGMsj722OMukqG9aAOOoYxsX7PmAy+TWZILfSCkxTEPPviQ7du3z668crQPCgwsAwYM9DbZsmWzX8fi4r62dOkS69Onjz300GctOzvr1IwJK7Ta80UwH3/8cy6CGRmZ9vjjsbKR16pVq+yuu+50wVy7dm17/YO1GsrOvoyMdBdURP/aa6+xCRMmuNgDIr558yZ79NHHfHBG0MmLQYZz77rrLrvhhhu8PzIQcB0ZqAP0mWuuucYHnfz8bl5X+iYDMgJN+blHKHN+fr6NGjXSj6dsXMPKytjsDD74YLVdd931dvfdd3s7B1avft8efPABmzhxordfoiKh/ghgUd133/1WVFTU6XOwdrCSuaGwIrk56PBNTc0+zUQkuCEQSQQBQ72iotw2btxo9fUNfnMXFhbapEkT3foMjBgx0tPmOM7p3r27Hzd58hQ/FsuN/aTNZ3Jykgsi64Pki/XDdoSe71jdWIFjxsQsG47FIurVq6elpcUmYvGTiKSkZF9CYCDgH2kh7KSDdXTjjTe6ZR3bl+o3NWKK5Uhd+/Qp9kEKCz+Qk5Pr1i9ignbHyo0FnnzGbAYLjXojVDt27PRBA4sYQltyTnp6mh9XVVXp7R4rd5KnS7kYPI4dK/W2AwYNBJQljIqKSh8waZvp06e1W/2nau//03akRbvwHYFkcM3MzPD8GTTIgxkSVj5iT9oNDY0+uGzbtt22bdt2aqaAiG72wZaBjO8MPOHabt68xbZv32bFxX3a2zu+XSg714qB6LbbbvM25/pwXDiGuqenZ3gZKBcWLf2L9kJY6Rvbt+/wwRKLmn3xM8cwGNOmzAhIH8s/zDLYHqs7bZJkzc0YJIf9GER39+5d7f0Iq5u6s518Anl53byue/bstqKi3paoJLUl8sLPJYamjt305S6i3DxYryxhYHXs2rXTLTGEhOkzU3FuGj5ZsmD5gMGBm+rEiRM2ZswY7+BBcLGkx4+/ql0IOI50Oe7kyRM2aNBgX+McPHiwWzrkjRAyzZ04cZL/zaDDjc80liUKptuIGnlgifEPoeBmwors3bu372PNlO0nTpS5OFNmRA2xYh/TeixoykuZEFIEiIEJgcayitX/Sj+fG5v0sbQ4FisUS2zQoEFusdIOlJPzvv/979uUKVNswID+LqYs0XAOSzbkwYBIuSlPbP14rQ8CTPcRP+qCFci1YUbAdvKnXrt37/Y0KDftRrswYHLNGIg4hvJQdwYe0qK+WL+UY8SIES50pMs1YxmDulA28mPALizsYb16FflsqaAg3y1wBh1EkPbgutKeLIvRBlyjmLWcYePGjXPrm2vLNaV9ghhTZ8rPNgZGZkPkR58K/YZ+wW8N1JnZCGWknVhioDwsmbBMwgyMARALn75F+5AWx1AO6kcfxJKm7s8++6xNnjy5/dqyhEU7hrbAOOnRo7vPcOiv9G3aC9GmbRlAgIFy7dq1PnBQf+qWiEioExhEEvFEgC5XEF7WOqdMuTphb+KowSCzdesWH/w7+9uNOD8SaiGEiDhaoxZCiIgjoRZCiIgjoRZCiIgjoRZCiIgjz8TLnPhnYrsyvQ+T5sWc+1HyOVs6erqga+iq6yK6FlnUl8HjZx3jHwTYtnz5cn/OtquIuYO/7WnyiNXFPGYfc4qZe94AR/HMnz/Pn6v9KGXlGdvgSfhxEeJenCtw0IX2R53g1EM9eJaeWDSdgeNx6ydWRzg/wDPcPE/Os9k8Aio+GrKoLyHBMQQnC+Js0ImDOy0OENwk/GMfzgV43OGhRkyMJ554woUAxwOcE3BawFkFBwocGhBWHB5wBMGZoaM1FCL34TGGkwE3J043lAVnCJxQyJO0SRdnBTzuKCN/4wCB+/uhQ4fbn7smPfbjzEI9SHvTps0eb4R0qQvp4RgSykM5OBYnEW5unn/GmQMX5+HDr3CPvvgy4wxx8mS5x4jAaSI8Kx1EEdf7/v37+7EhD4SBvHE+4ZPzcPTBKYV6UB/qzHnx5cIhA8+54DEa9uF48/TTT9utt97qzh84heAIQpuRF23+s5894y7dnFtSsseddnAcIT2OYXt8XlxH6s215lqQFvvjjwv15Loy+IU+QvvF8inxY2nj+GfIyQ9Xb9qd+uL0Qn/hHMpOmcgXx6uyshNepxDcCSepnj0L2+OS4MRDO+Fw0rFPUbYFC+a723ePHoXuPo73IOc0NBDnJN1jfowbN94dX6gHDjO9e5/p4Ui+BBKLebIWnnKsKfK64jyzb98+d68fOnSYOzklIolZ608IOjY3CWJNpyVuBK7X3DwEOiIGCDcHnZZtCC43IZ6K3KjHjh31GxwPMFzOY514v98UBPiZNGmS3xhf/eqfuRDFg9g8++zvbdSoUfb++++75+PevSWWlZXtghJubMQDEIMNG9a7ZyCedDFrqc33E7AJ7zTKidsyN9qSJYs9wA5l5JjZs19z13E81P7oj/6o/cakDSgHLu+0BZbX4sWLvS78TQCgcCxxL5577nkbM2a0B52aOvUa9zgMEdXIH8F54YUX7E//9E/PEHnag7gafN5xxx0uhswUEJWFCxfZTTfd9AfXh/wJQoTn4pe+9OUzrhsDCdb/0qVLfXDALZt64mF39dVXu2cj+xmgqqtrPEgVcTx+/vOfeSyNeLi2L730kgsncTVwT2dgQJAYkGmLAC72WLjUjfxw7iFuCNeHdIiLgZt8vNMSwrxo0SJ323///VUeqZH2og6LFy/xfsIxiCxOKYBXIG7qCDJ1xECYOfNlD5703nsr7I/+6I/P2qcZqDdu3OQD+4oVyz241/z5893TEi9L2p12IdgX3wlD8LWvfc3LDFxLwgYQ92Pfvr02bdp093Z86qmn7JVXZnkMkjWn6kg70daJiJY+LiFYzohLZWWFWx2I9DXXTLXRo6+0w4ePnLppy92SJpgOYoJQ9+3bzy2MnTt3+XncoPy77rrrPNANwkAcCwLeEEchhNSMh5sEIYq58PZ3t3JuSGJWIIwEvcENGwsW0GUGA4SHaGYcz82Nizhl4AabMmWybw9BqRAQ3IMRa6wwbrIQ6jLAgIOwU3ZmAgxaiBzxQ7p373FGmbkxiSeBWODyTZ4EbwrhRrmxWXIgols8WI/Dhw+zt99+261F3KQRQcSAPBBClkzi24jv1A8Lk/rFw6CHdc7MgbQJHkSQIq4X9eO6EquE/dSdpRiElLqOGnWlD1jxFiThACgTeQaXfeKH0GakFw+DKQGpcnNzfEZAACmCMpEPgyvXvuM51BmRXL8+Fu2QtsJdm+vJ7ILriTs57cB3rFgMBYwE/mYwpx/Q77hOISzp2ZZ2GDgJ3MR1v+qqCR5xj7oSpIkQCaFd6COEE6Dvd1wKOXjwkF177bXuDo8xEMKs0q4xg6bWjYgQnyURkVBfQuh4TOewjrKyYnEX3njjDbdmEK0jR46e2kdchiwXBf7GosayClHziP7GjYG1yPSXZQEix2GBhEBCWCnxa4ZM1cmDm4ipLTf7l7/8FbeYKQfrxfxDWLCymUpzU5MPlhWijqgSeIf9WDh8Z3rLDcTfs2bNtPLyCh8IuMEIzTpkyOAzpqscR6AdjsVCJM4IAYnY3nHJJhY/o8JeffUVH7yYxmNdvvfeey4QBBxCPGJxU05P/UkD0SDyGgMQ7UqytDkCikgxIHYUaizl0P4dIT8sS/aHaxIbINP9b9qAULWEAiXgFW1EvUmr45IB4kWZ2c8SCeVJTk7xz45u8NSRbexDrMIgzMCOADOgd1wOYKDh2rFkxEDK8fQZBgLO53oyW0GYCYWK6FNellYIocvyB+FVQ9ljUfGO2JIlS87IJ8RDYTZFH+Z8BgYiMhKrg3ahb9IuhDFl9kU+HduDWRFxwZctW+7twHWi7zCAM2ikpKR6/UkrUZEL+SWETh0Cs9PxXn99jgdhQgwR0vByAG4MjuUYOjXiws0Ygq2zn78RLzo+f7OddU7EjRi+CAGWTPy6KMsf3MRYOwg752GtMD3mpqIs3AzczAg+abI9rMkGS5bzQiS7MBiEwPHsQySw8ki7o4iGoPLUieM4PhyLUHdco2Z5CEsYsf7jP/4TryPiSPmoG2VDwOLXwePbOsSVZj09BCbiO9vjRTT8IBjq13FfCLpPmWl76k46fGcb9Q9R+xgEuC4hCFJH4Q95sRTB9WKZgTqFMLXxwhvW9InCx2yAOlEuBgby5zzCpDJoxhsEtAvHhPpSPtqXepAva8pcZ+rLNaecpMUMivrwL7QFZQ1hdNkeXzbSIvofAwazjNAH4gdBBjX2UQ/2dbxWHMtMjR8mCapF8KX4vlR5KtxsiGudiEioPyFir3A67tYlN0FXpssNxU3X2cerEG9ugBAaNGqEH1E7htkUny6i3g8/SSTUQggRcbRGLYQQEUdCLYQQEUdCLYQQEUdCLYQQEUdCnWDwBEVwWOlqSJPH58KbpS+W4CTSWcLjdJ0pV3DaOBc8V/xR4o5cDJSnM2XqzNM9nbmOnT029iRS2Xljc4Syf9j9F8r/o5z/aUZCfQkJnZCbM75Txv87374P869j3jzri5ND/LaOecef09myBsEhLgk3e2fq1PF83K5xCDrbvrPVhefFcc443zHEJwGcXxhEzlWGlStXudfouep+Me1woWuyfv16H8zwjuzYVudK42zlwgEFsT9//WPbli1b5gNRaJOz5UXfwBHqbH0jbAtlP1c5ed6Zt4+f69pdqF/hvn++NkxUEvPp8U+IWBCbBR7fAlddnBJwba6vr7MHH3zIYxzg0UfcBASL44jXQGwNvOBw0ebN1Nu3b7O77rrbbxrciwmCRPwObrT77rvP8+B7r1497fbb7/Bnj7m5Zs+e7c4Y/OO5ZCLd0fnvvvseFw22c3x46/iyZUvdo4w3aSMIlI10Hn74YRfJPXv22NSpU3077um4EuNEgSMPHosPPPCAR/fDA5H4Eggl+cyYMd295ghehAchz86uXPmee1kSy2Hp0iW2e/ced4XGxX3x4kXu7RZzMTb3kMQ9HI+6W2651WNMcC7u9ldfPdXriwByDG17/fU3uIPPtm3b3LWaeBa4Lcfc0Pd5HrhQh8GGulMGYmRwnXAJj73pe5B7btJunEsQJq4T+2+++WZ/izptQvAgPC7feuttd/ghlgqBmwj61NzcZL/85S+9zSkLbut4nOKtSLAo8mbAwLWcN8BzLa+8cpRNmDDR68U+rjXtRHlxyWaAozzEQsFZhOO4Jogz1+Oee+7xY7kOtCXH3nDDjV5X2h3vT2J04MSDuzv9imf8jx8vs9TUFO+bXCOcUohdQryU/v0H+HHUD49HgibFZmrEWZlpf/qnX3XnHPorrurEW8HxCWudeB+hT9B2XCvSpy8988wz9sgjj7rDC+kRRmHFihWePn2T/BIRWdSXEG6gAQP6uzca3oCIC1HFCGBDJDk8FBFXXKRxxb322mtczLmBcYrBfZrOTTp4eyFC99//gKdHgB86NrEhOP6uu+60bdu2+5QXEFK8yO64406/KQhNyY2NizLu4YgMsR2C8w032dy577pgEDRp9+5dNnbsGI/JQfoILoLNTUqsDWJR4FlG2Rg4cFpAFAgiRPlIA6FlPxHWEGE+EfKFCxfaPffc69u4QdetW2+PPvqoB5/iRmdwwH08QFrTp0/3suO5iFBRTizMsOxCOogUYoqwIVw4AS1fvsJWrXrfhQvBiNVvkXsHxi+prF27zuuHZx8DI4GVcJNmgMMTj3gfpEu7IUSIHANDKMf8+Qts4sQJPrgysLCswzUgIiEDAwMAsS+4FggRrvxYuqT5yCOPeN8gbACiSrsEEDa8AAkDEGJ9MOjdfvvt7dYw0G8o+1VXXeUxP7hmeChi8TJwMcPgetGGDA70E+rGEhBREEmbNiZtBlgoKOjuZaffzJkz20UTIwLPxpkzZ7n7PNcYt/WBAwf6tacuDIYIP+CKHt8niJfCIEHsE/oe6TMIsZ3Bj4Hp+PHS9hlIoiKhvoRg5RLbAxfaML3jBuEflgYWLzEZuGEJsoTLLFYI7tvccNz0xOMg4llGRqY1Njb5jcf0H6EnRgKCg+stLruIavx0kc5+eq2SNVLybvAb+r777vc0wvQfQWU7VuRnP/tZt5Dz8wvcNZqyYoGSFlbi22+/40GTYhZprE7B/Zz64Mp+7733+ECDqFM+ouvhHhym4RxP0CAIaQMDGa7tWGnxhOBOCB83+IABA+2zn334DBdjLNbQ1oCIMgNBtLB4Q/2IcofQxhPKQBs1N7f4d9qE4+LTvfPOO92qRvRCORB4rhuCjtDynUGUCHKkwb/YkoV5sCTYsGGjx9egbcNyxogRV9iECRPs5Zdfbr+OBM5CIFlioq2BkAAhaFU8CCjljF9z5hqGspPP8uXLfAYX4qAcPXrM+1DsenezzMwMD4QF1D02O2tyl/Xi4j7e5rHrHHM1Zz/nh+BZoS7E6wj9KvQJLH0GT/o3/TlGrG2IdUP5iQ9SUNDdrrnm2lOhdxMTCfUlBIEaNoz4w1UuQHRarB06NcGDsCawesaMGeviRGdlG9Yk1kTsBQLmUeMQGW5yrEIsOm4WAvyEKTPiQBzgEGcjWLAsr7DccNttt9v69RustPS4Ry4jrjUChIUMWMRMOwkxyTSVGysvL9cD71AebpzXX3/dBQyxQ5BYImHqjRXMDYjVidCQBpHwCMfKNBtLkvoz5aWshAxduHCBLw+QNssXc+bM8Tywivfs2e3bApQL6xZhZCbB8gf1YtAIcPMT35rBgMA+tCNLOpRp6tSrva1uuulGLxvlwAJkiQI4B6uR+iEc48eP8/JgiSJOXEPah/pu3brNryP1uPXW27wcCDdLVswIWHNFYHbs2O7ncH1YxsASZakG4aRMHIMAY11irdIfGFixbMNyDrAEg5gSRY/rSz2ximkL2j5APiE2OfHHWdKgfkQVRKgJM4sgEu6WAZSBHQGm/1EurGICK4UgXQHKxYzn3nvv9bCoWMvU97HHHvf8GQBw9WdpiXpRF4wMLGjo3bt3e59g1kZfZ4ZAf6BcWOQsqVBeBt3Jk6f4YEHI3Yv5ofnThlzILzHxVg9Tfjo3QhsikXXkXNsvhvjgQufaH7/vQsdfTL5d+QNQZ8rV2bKfr707c35nynqxaZzvnIupV2eOA5aMWHJgMOK3jbNFDfww6Z7v/ItN41x9M9GQUH9ChF+xw1RYiE+iD7I8EZY01A+ji4RaCCEijtaohRAi4kiohRAi4kiohRAi4kiohRAi4kioIwrPoxIzgWencZRI5DgHQiQ6EupLCGKLY0jsZaWN7pobPLdweeZf8FbkxbU8Z403I2634YWlIeIanm54JUrAhfj0o6BMlxDckn/0ox95PA8CLuEFiPcXbsFLlix2McbLbv78+e6tRWwIPPOIU4E7cuyzwUaOHOFeYXht4dElhPh0I4v6EsLqBa7VxEcgNgMxJnBfJhJdXV29OxzgkovVTBCgPn16u0MCVjUu4rGoaCXu7ltTUxuXrpZFhPg0I6G+xBA2Ek8whJdP/vXtW2yFhT08ahhxKRBmxJvIZYAOE+GN+A3EPyDOBzERiDy3desWj3omhPj0Is/ESwhrywQkIkASoS/DJ0FsCOCDJU0wJoIgEWCIQDYEsWEtm3MRZLYR3Ic4xgTP4RyCCBGcRwjx6URCLYQQEUdLH0IIEXEk1EIIEXEk1EIIEXEk1EIIEXEk1EIIEXEk1J14C0v411XggRheuNoZyDu8DLez6Ye3cYdz+Xf6xbbW/uLRi0kXOB43+PBCWtK82DQ61q2r27dj+rwL8FI6BcXybL6oPEN7cl0upm9cbnyc1/rTTEIL9YVEgpd2/uY3v/E3YPPSz65i6dIl7oHYWXjDNW8vP3GizMt0oY7+1ltvuRs6cOyvfvUrf2Hod7/7nfY3Ur///vv24osv2u9//zsrKSnpVDlIi0BRr732mruyf+tb/2y///3v7cc//nd3db9YiHkyc+ZMe/75571eHxevvvqaOxF1JWVlx70NzsVrr73qsVs6CwPezJkv26pVq2z79m0WNbj2PMcf/0bzD8PWrVs9fIK4OBI61kdLm9nLe8rthuJc652V+gfvjMPphBd+du/ewxYvXmw33HCDewziIcjbuLds2eIu4LxlGddwBJW3PvN26lWrVlr//gP8lfe8QRzHFd5UTR6VlVUubO+/v8oDLfG2bfIilsc111zjDi3cuPPnz7PGxiYbNGig5eRk2y9+8Us7frzUvvrVP3PhbWpqtNtvv8N27NjhDjC33367O8QQ0Im0EJOwDQ9I3g7+wgsvuPcjb0Kn7EOGDPY3WHMDUZ7hw4f7W8oXLFhgFRXl7lRDeYniR1yRJUuWeIySDRs2eH15MzdOO7BgwXyvz/TpM9xrsq2t1d/MTZuR3x133OleleHGR6R52zRvAE9JiZWPt25TPt5ITZ0QWF7+S11Il2MoN29g503avOUb8vJ4G/dIP583k+/atcsOHjzgb/ymbRkceeM3b8CmPMwKpk+f7tcX8WHgYhvXiWtJnrz4tVu3fBdcYqvQNqSN5fuLX/zC30bOG9FpB972ztvcGXA2bdrk5Zgx4zq/9lj006ZNd6cm3sbOceRBH0CY09JSbezYcd52PXoUepwXrkdJyR5/0zjeqLytnu84R3Hee++952UcNmyovffeSq8DfYd60z680RzvVfob127FihX+lnnezE6e6elpNmXK1V4v6ke9Cf7F2+YPHTpo1dU1NnXqVDdQGJwHDx5kv/vd77y/0Q68Fby4uK/3D8pCubhuOGdRD/oIZWEbswuOoQ/n5+e7Vy7Xkf04avF2dN6YTv6xa13mgxVvQOfYkpIS7wfXXXdde/9JNBLaom5pa7Pvrj9mt762w17YVe5/dwRBxH177NgxbvEgeHPnvmN79uxxMevXr5+98sorPl2ls7/77ly/cQmoRKdEkNj+1ltvuhAE2J6dneM3xIoVy+2dd952QeWGh5UrV9rixUv8htuzp8RvfjwQ8/LyjPGEDswNt2jRIvvlL3/pQhc/0HAzHzhw0GbNesVFk2h73LCZmRn27//+f622Fgu4zfNB1H/wgx9Yv379XYAYIBhoEOOjR4/Zz372THuaiOqsWbOsqqrKRY5BZdGihd5Or776qt+w3PiIDaI+d+5cv8GzsrLPbPuWFg/fGt7CnpKSam+//bb/TXsiIAxAiCxu8tzEO3Zst9dff93bc968+e7l+c47c31A5Bph4ZInAs0MhDSx/hsbG7wd165d49Yc12X//n22ZctmLwuzm82bN7l4ErGQ64dYrF+/wfbv3+8WPwMHMytEmmuEoNIW7777rrcD7bF06VIv++DBg/1a0ye4pggS1+/dd+f5IE8fAtqNNkAIEa54EGfiwbz55hueLm8KDyLF9SJN0l64cFG7lcssiuMRvnnz3m2PrsgASnoFBd3tjTded3EsKzthK1e+Z7Nnz/YQBvQlri9tumbNWh+k6N+vvDLLRoy4wq8f+Q0cOMDz4hqQVmxAKXGDJPD22295fXbv3uV/M3gxoBYW9rTdu/f4NeA+6dmTvr3b82Z29cEHq32QePHFF7y9Gci5Lnv3lvgAQDslKgkt1JCSZNYzM9UKM1PsbO8CnzhxglsyuHVjKcC1105z0eSmGzRokFtMCBUdkxsZi7pnz14uvnS+ffv2tq/lBhAzIuY1NNR7ZLx77/2MW7QMAMAyB6I3ZswYt364uYuL+1jv3n28LMuWLfUblIh8TzzxhAvMtm2np8xDhgy14cOHnZp+xwYgblCs5djbp2Nr2EAZqAPWLTckAaIQL2YHzAK4CW+++WZbs+YDF0vOZRv/GKgA65sbGcuQm23Five8fIjLww8/bC+99KJPnYN4UJ/U1DS32LC4GEiys7PcEmMQ4lzKz9/Upbi42OrrG7zdse64kWlrvuOKD6SDKBOwitkGAnrttdf4jGHw4CE+MzpwYL+9994Kn6nQ7uFaIG5cy1jwq1hasXZpdhFmGzMqRJp/eXm53j6VlRU+46GcfKfcCDUDA/VlwCMvrsHdd9/tfYJBAPhkcB09eoxf60Bzc4uLOOJFnW+88UbvD/QP2g6rnhkdgxF9gQEDixSrk5kFVjHtFK4xbcv+WBkr/fh+/fq6kVBU1MuGDh3mAzh9jX5KWfv0KXaBpj4MhLQ/wcBoQ2Yk+/cf8GOHDh3i6b355pvevxkEMUAYKJjNUD/uk0mTJtucObPbfydhZrFhw0a32jmGdqautBXXj/5x/fXXW3Jyis8CKHt9/cUvr31aSGihTklKsr+/uq+9dvcwu6VfniV3WPpAOOhkTLkOHjzkHQdLDGuDjoSl99xzz3knRJC4CbBajh494jch02ZEory8wi2wYEl1717g00AMeMSCzo11hwWIuADTUizDH//4x5aWlu4WDFNtLAysITo7wsuNuXHjBhdNlgOAQYJpM8s1t9xyiw0YMLB9ivn00z/xaXjfvv1c9LkBEDpCrX7ve9/1aS3pIFgMTIgzIoc4kQeiRnoI5aBBg/04hAmBbmpqtn/7t+/5Jzc4wo0lxwDCjc6xiAs3NDcnwvXSSy/5GjrWXEZGpv32t7/16TTtjjAwg8jNzfHy02YMCM8++5xvwxLHSiMtxIY195aWVh88EJdYJEIGle7ta8aUGVHgUiO2QPmxJIOQjBw5ygX10KHDfq1oG6xGhC3MWmgD2hcxmj17josX7UrZWbdvbW3za4/w0HYMSix9IKC0A0ybNs0HKmYLbCsq6m35+dQzx49H+BHSnTt3eD+hPYABgHM4luUOrHqEnfR69y5qF0dEju+UHyFlVsf3V199xWdLkydP9jw5hkGH8iLWlBerlxC8DJL8TsOMgGtIPizVIP60PTMuLGbKEtqGpUD2h8GHOnItuIb0IfKh/2ZlZdq6devdsMGyjy39xIyDkpK93idCe+Tm5rbXPxFJ6Fgfoeod16YDwQJmP98RWm5kOjJTcpYssITDjcc+hInjsfbCdr6znfNDWsGKANILT2aEcKbhyYGwP1ih5MExYQrONkSR/eHc8DefTOf55DjS53soV/ghNQwgpB2m/0xNOQ4rknXoUIb4fEmP76QZPkMdQjtxHOfELOjU9mPPrGObC1NInzKcrVuS9q9//Wt77LHHfGAMZeAzpBVfNv6mLOHJF/JkfyhP+BfaCqFg2eexxx4/o56UD8GNv4ahrOFasD20UXw54vMKSxShj8Q/HUI6oYzh2oS2Ao6jXcJ5oR/E9yPSDe0RPuPL2rFMod/FHw+xmVqs73S8JqFMoU/H93W+82/16vd9UKPvxJc3HM+AyXVkVoDR8/jjn2tPN1yP+PTjCWVMNBJaqD8KdCym8VhIn8a6YRFxU2DtR+XmoKsyvcfCOtfg+lFobW3xZR+u6ceRfqKAJYzIBoPgbNeR30v4h4V9NkEWZyKhFkKIiBMNU0kIIcQ5kVALIUTEkVALIUTEkVALIUTEkVCfB54wwNnhUsUbCcGOLgWdzY/Hts4VXySUm2OCI0vH4/ibZ8p53OpC5QjxJMLjZkKIGBLq80Asj3Xr1rU/PxpEKF6g4r+HY+Ij1sWfG78/fOcxOJwJEDriMcSn2TGdkHf8to7/znZ8fH5hH5/kFzwmz1Y2YD9uxGfL9/DhQx6Pg8EMhw/iW+CI0THYFc4W1PNc+YRy8FgXDjZnq6cQiUxCB2W6EEEsiEWAVxmusjfccKNvwwsOd2UC/RCYJ0SUI9APQZGIS4DrOSKMNVlY2MM9CgmShGjhYYjnFp5zOJj88R//sdXV1XqshptuutljcPTv38/WrFnjz/Y++uij7rWIxxgeZjzbjCccLuekSUwNvNTeeecdF73bbrvNPcn4PmfOHJ8dsJ/YDATrIV4EwlhTU21vvfW2B3siOBSef1jQV1wxwq6++up2QcXBh9gP1JGy4F2GR9zevfvs85//vKeDazbPxBJ8CS9NXLKJ18HzybQBXol4bd588y3eRgyCvHUdixonCASdMnEOrt54vXEuDhHkJ0SiIou6E+CWi8s1EdaCNUjEuPvuu99dY4kMhiv2vffe66LLvokTJ7pQ4tV20003uhDNmDHDBRzXZwIgrV79gbvPTp48yd24SSspKdkjpRHUplcvYokMdnELYUCXLVvmgwWRy1hSwFWcgQMXZ+KKIKiIOLFHYNeune6CjPs1oo8r+ne/+12PF0EMksOHj/hSA+7AnEcAIOJUELAnPjYJgwTCS92ILxGLUTHapk271t28SYeYFXgx0j5Y1jg0YHV7R0tO9ngTuIkTeW3p0mX28MOPtA8alJ02uP/++23hwgUeSwLXfYIRhRgrQiQqEuoLgEVHVDhiDcR7T2HhIV5YkLi+Mv0PgYWIsYClzfGch7szMTn4JFAOcScQzuTkmEszFmVYl0XgsTyx3onHQIAj8gnT/9TUWF5EEwvu4qzrYqESE4RYCYgnwW5ix6e59xcxKQisRNwF4kCEwEBYuljlBH4iRgblxZp+8MGHzvBIxC2YfIjPQFli7tipLsaUgaahLljulBHLnPgZBBgC8iNORIgoyABGHbC8QzkJhUobxto1xT0QicJHmzGofNRYyEJcrkiozwMWHgGJpk69xsNRsnSAQCFgBLdBmAhsgyiyjcBK99xzj8fgRTAJQoOViWgTy5ltxNhliQHR4xMLFxFHQGNBcoo87jJpjho10stB9DDElrxZFkFYEWbyx7pmfZi0pkyZ7EssRCUjGh4QTwHhxUpHRBHKr33tax5djmBSCPWAAf39WJYiEHM+EUzADXjy5Cm+RIFQEgOaNELaxMBgkCGQDtbxsWNH3TqnvFVVlTZu3FjfR1sSOIl2oi3uvvsenx0g4NSb9mTQY6mImcn48Ve5K/eYMWM9GA9LLJ/mN58IcT7kQn4eovYjVghwA8RrRthYOjnf8ReqB2veLNeQFgHtWd44W37nK8ulRDEhRCIiof6URv67mHTi05IQChE9JNRCCBFxtEYthBARR0IthBARR0IthBARR56JEf6xr6t+MOxsWmd7iuPj+HExPp+LTb/juV3ZRpe6LonIJ329LldkUV8AnDKI+XExj6JxLM8i4513Lnh++ULBh3iWGW/IrgDvv+3bT7+l/GxQHt5lN3v2a/b6669/bMGR8M7kxbCdee1WxzKwjZfHhu08Q467/ScFj0niyBMFaDM8YLvieXP6J85MF/sIJs/O4w17rvPo04RkEBdHQlvULW1ttuVkvV2Rn2HpyX/4aBqdbcmSxe6Y8V//63/zF3wieLiA4zXHd5xacLnm7eS8KZr4HcTXwK0a5w5iY+DUEV5gijs0zyoTN4PsiHuBswsCxHaeZ+ZYnm8+fPiwBz3irc7khRMJbxKnnHgzxt7snOdv2eZ89nMj8Mk+youjCceXlR23qqpqd0bBBRzPR9i/f7+7v+NcQl5z575j3/jGN2zXrt3uFLNjx3Z/ozb1Ig/Sw42cN4xTL7wreVs1ZeVt3tRx165dnj7HUB5uetqJfPF+JI4H9aNe5EmdcPzBbR5nGspD/f75n//JbrzxRne4wVWet7rjOYnIjx59pfXpU2x79lDOJhsx4gp3o+fN6lwP0uIz5sYfqydekCNHjnRnI67DFVdc4V6QlItrhbcpAzPXFqciHIHwvuQ8PC1xPApeonhdcix5cu3x/IzVpdDfOB9z+W+z7OwcPxb3eepJWXGr5y3ptAHtgeMSbclxlAmxZV9oz+Cyz3HEQuEc2pu0OB5PVTw3Sf873/mOfeELT/i59BkcqGgH6oyIEpaAstHX6B+0E45YHINTV0VFpbc/gx8Bsv78z//campqPTQBecW/I5R2IC08Sgl1EPou3qj0aerBNq4BdaC9qA/XjOuApy73R3j5MteW82h/2pQ3pHNMeflJ77uhPRKRhBbq5tY2+6sl+60oK83+26TeNq5H1hliHbupDrvXH8GMuEn4G6/BpUuXeKfkZkBwDhw46DcL8T+WLFliX/7yl+2Xv/yF3XrrbfarX/3KHnroIe+UWNlbt25xa4UOjaAuWDDfxZYb9frrr/dtr78+xz3y2tpa3brlpqKDf/nLX/EO/corr3iHJkbGI4886lbKww8/bLNmzXInGPJBjPDy48YD0kKA8B7cuHGDxxXhxkPY+aRuiPHJk+W2bNlSz5NgVFu3brMvf/lLHnjqP/2n/2zf+9737L777rNnnvmpB0yinXB1Zz9CRRCpu+66y4UaIaOsCC6BqP7iL77W3r7EIyHi3qpVK92jEwG+555YeWmfPXtKrGfPXn6zEqiJAfOv//obHkDq7bff8ZuZwFGIwQcfrPG81q1bazfeeJPHNbn77rtdJBDaf/qnb7oXJ8f+5Cc/cbHDGn/yySc99si6dett2LBhLmgMJAj+c8/tcyegN9543dtnxozrvI25Tt/85jfd3R/Xe6A9KSNxXz73uc/bv//7v9sjjzziwkLbvPjiC/b3f/8PPgB9+9v/6gL4zDPP2D/8Q2zbP/7j/3EPUsTthRdecCH73e9+53X4+c9/Zo8++pgPohgJV1891dsKN3vKRP9hIKDdaANmgLzdG29O+sdXv/pn9s///M82ZMhgHwCIW1Nc3Mevb21tnZ/z6KOPeJmHDRvu7U2foD7MGF544UUbOHCARzn86le/2n6P0McIr8AAsW/ffg9OhtDjXXriRJkPAKT90EOf9T48ZswY7xOwaNEijxXD4MY14T7ifGaQ9fV1PnDPnfuux8d566032wcn3myeiCT80kddS6stOlxlS47UWEuH2RqdDNHk5kW4hg8f5tYL1hk3A8KNJcIn4sF3Ohju2+GV93ziWk0n5QbHqjlx4qRbVLiaB0sDawlrDQ4ePOCu04gbMTAQdiLrYQ0jtkxtsX5uueUWv7FiVkxT+yfiSmfHOo2fJGAZIxrcTFhw48eP8/phLQHl4YanrET1o17Uffjw4R7EKSzXYGkTA4UB6s4777QjRw5bdXWNt8HEiZPcYkREgUGDm5blFFzwA8yMKQP/qD/igXBwk1M/bnIsKFzaEe1jx0pdxPiOWD311FM+cwnTc/LGGuPf0KFDvHy0A/toVyxeRGDbtm1+LDFEcnNzvCxjx47zmQPiR923b9/h9QzXctKkSXbXXXf7ABeuT58+vV3kKUusLifs+PEyHxRoN9z/CTDFOfQZ9mE5MtNiQPj1r3/l/SlmMZZ7XagT14x2+OIXv+gRCbFWaWcGMmLIPPXUH3mfYR+iR16c/5WvPOlRCxFrBI8+xkDNvoaGeuvXr6899tjj3s5EPyQyIbFkMB769i32+nLuE098wduQdkBIMzOzvA4MOJQ93kGKdoi9rT7JY9zErivRFpu93zJgMuvhPmImRn/mHon1xRa/zuTLtab+tD/hBRggYte5p7fZuHHj7Nprp3k/SlQSWqiTk5LsC8N72Lz7Rth/GN3TUpNPqxodDtH4/Oe/YI8//rh3VCK6cfMGIeXGIYodn0wPEYRYfOc2t4gRrzClRuSwuLHM+JulACw5Oi6WMRYMywtYK4g4liGWEgLOTcMnnZ0bhzQpB+FLN23a6Dc7YkCYUGJvEF8kdkyRLxcAU3jgpqEMLFdUVlZ5bJDTEe6SLCsru/3GQxhJD4sH4ebGevrpn3hZuBGZ1lOvHTt2uvBRD25ypv4IAZAXInno0GGPc0L9aEvEiKh+LA8wc+CGR/SIFBjWWIuKernAY9UhmMwksCIRl+9///t+PO1HetzclJtjECmm+CE2NhYrZSYsLdYf55A/EQwBoQhLO2PHjnWBQkgQt5wcgmpl+fG0GQwYMNDr86//+m0XTerOzIC8yJ/ycB4ig+jzdxgUgMGHQQNrGBBcruG3v/0vnh5t+oMf/MDzoWy0MxDIi2tBfwh5MIhSn+9851/tgw8+cKEn6BeDeywoWJ5fe9Lg3JAG14oBCKGkrYhHw7XiWPox/ZFZJP2K5ROuN2F32cZ9EdoB8eZcBDnEwSE4GH08tiyW7m2LxTxnzmyvH/uZsWBVM6uiLtSJ60x9Ro4c5ccwcNAuaWnplpISSzdRSWjPxPP9Ys8+OmmImodoQRAUOjvrc3QyprsIAdNNxIi1PG56On4Qb77H1mv5nuMdEKuYqT1pczw3BB07ZiHG1k05jk6PZctnsEi4mbGE582b55Y1Nx75kB/fQ1hU0gxr5eFNLPzjGPLBmkX0qQfCEta4+fzNb37tN+7u3bvsiSe+6FNnBJtjQ70YcBBnrDbKyqDFudyc1AVeffUVnxp//etf97ZkdsCyD+XC6gvtRH3CwAUcV15e4eVhjZ1BBGHAmmR5JTZjiD35EepCOpQDq5m2Je0QdpVrxLXjOP7RNmEWw7VGtBBNhJ32pm6xiIctpwar2Fo76VFu2pEy0K6n65Ll1zfsI5/a2hoXe9KmvAyW//Iv/2Lf+ta32te86QOh3bAm+c51AdKKtfMJH3jJi7ZieyymeZPPTDiebbQbZQjtQR8mTdqVfsZn+JvlBK4TAwL7OJa2oOy0GWXmOlMeBjl+OCWYF/mGNg9iTv+kDWNv+2n164XFTxtSvzADpbxso425Pvxjhrp3b4kvHbKffZSFMjY3x/oc5Qh9I9FIaKHuKuiEWFR0PiwCbupLQRCgYJF1ddoIMJYqNy2DD0L3YdJBmBBYblI9lmXerggrSyCXU3uE2Ql9ravLjWFBugxI4g+RUAshRMRJ6DVqIYS4HJBQCyFExJFQCyFExJFQCyFExJFQXyQ8isUv3x/3wzLhF/az5cO22CNQbV2abpShvDye1dlyxztmdOZYHgHj8bTLrV1EYiChvkhwwSbg0sdBiPGBaPBIG7EWYm/yPjMwEc+Z4pL8YYLv8PwtLu6XGwwuL730UqeFlMBPxCTpLDgLvfvuu50OREXb49WpF+6KS0FCx/q4EDyjzIP4PNCPW24sBkRvFw2cG3jrOJ+45eKajft0fn43W758+SnPr8kegwMvN+I4EHcBr71p06a50wLxKPDIQvxD8KAf/vAH7grMM8cEpfnhD39ot99+uzsb8CxzLPBQb3dVxmmBvBAXYoTwrHIAR4bVq9/3NMgDcY698TzP0+E5WASfmCU4n+DeTcwN0sRjjnxwyMFRA+sdp5MJEya4cw/PAHM86SFUy5cvc29MniGnLNQdRwjeUk5b4YVHO+GiHosfss2PxRGEAE44CeG6jgsx5aO85IPzCHFWGLAOHz7irsxYvStWLHcvNcqAN14IsEQaBNHCUQJ3e55nX7t2rXvF4YUXYp4QlyPkS0wNri/twJvR8byjbNSf43Etpy2nTLna80H8GSw479/+7d/s85//nHs4EjcEJxDif3BthehK1KPOAxYT7uJZWZkuFqtXf+Cihts2nlMIy9Chw+yNN96wwYOHuJfgG2+86QKIQOAsgifXiBEj3a13/Pir3G0aYQ7pI1ocg6jgoYUrLvExOIapPsFz+BuRR6QQd0QacMHFSQCHlPnz57dbm3ySHi6+lIk8ib2BsCI0uGkDrsA4XyCgREvDXZxB580333Bhpyy8oRzX9aVLl3psCNoDF2sCLQFlYjDCvZjBCscf2gYXZkCoEU/iNeBWj6XLOXix0X5Tp17jQY/mzp3rAk/UvbfeessHB9yLqTcBl4idwSAwf/48d32n3QjytHDhAo8PwfmLFy9ybzgiwyHEzB5iLs75vh+ob3y+5EO7EwoABxQEl/KtXbvGrx/tmpGR6df4/fdXeXsj/MRS4dpw3V9++WV3zY6VraJLb1AhJNQXAKuTmx7B4oaOWaEEAqrz/bgoY8UR1AbrC8sZEcXaI+oa7soIAAKJCPI3cREQYNKaMmWKixKR8xALxDW4/EL4O2Ypxyxg3IMDBHjCdZd0g2t4mLoT6YxlEwLkDBw4yON/4MqNO3S8RyVejZxPugTBIb5GzBU904Mikf6gQQNdoBCh7t17+PGkw4CAuN55510eOQ+Bo31ICws7kJfXzdOhDtSbgQcvRwSQmQJxJLDASZd/WLhY3ljMtBciSGwIYm0QUIqIdZMnT3JLlhkKbU8sCAScOnBdsGppLyLOMVhQLqCN4vOlHpQF6z0WLuB0LGZcnqkjbXTddde5GzNtgrs1YyKzDa4b7cg1x5WftMN5QnQVsqjPA+6sTPm5gREEAvbMmfO63/gIFEF7YvF2+/pNzCdLJFiWxLZgG9YoQoAoP/fcsy4ULAkAyx3ErSA6WnDJRRSxDFn6QAQYKLCcx4wZ7WJIrA3EG5GaNm26W4dM2wkHicUaXlbQvXvBqfgPfTxuCFN5puaITgjUxBIDVuPMmTM9yBTW+jvvvOPhQLGsKXcsXnKqf3IMFvXMmS/7UgtlRtCwzGkf4jXv3r3HLWGEMMA5zz77rC+dECiKtkOQiaX9u9/91stBZDWsYwYuRJFAUbRhLN5Fks2ePdvPJd/SUuqzt718iDLtQfhZ2or1ZqxoBrhYjGOWp/p6WTrmG+Irs59rSSS90M5cJwJkEbOZa05ZmAmQBukyqDETufnmm3wmQdszUDODEKIrkQv5eTibVRSCnIdgQGf75JgQSSxs63gu/7Bcf/vb33jAoxDAiDXfcE58wCHOO1ve4ccsguCwPMH6cgjgHn98/He/8KfSCOezL/77ueoWzkEkw9/x9WVfKDv/EC6WDYjj3DFWSHxaHdunYxvGt0loI847W9uHNeL49gvndsy34zFnu8YhKNPZ2iL+2nS85pdTHA8RbSTUnyAhml1XBLmJ6rv7qCMDEpZolMolxOWEhFoIISKO1qiFECLiSKiFECLiSKiFECLiSKiFECLiSKg7EayHR8z4/LicGEgfh4+ugnLiHcjTFhc6bs+e3WfNm3144OHMwfPBiRjT4mICO3UmHTnBiA9LQgu1R6FriT2TezbwtPvZz35mr7wyy+bNe9e3IdjBKy+87j4IetiGh2L833wPz/7G/8MDjk/iS/BG8vhzwr5Qzo7pdPyM3weUF3fp+HzC9yC6eNDNmze//fnjEJ2OZ7L5/vOf/9xjnbz44ov+GCHHhLT4DN/JOz6y3dnyjN9H+mwL7dixXqEdzrcv1KPj9o7tFZ6DPtv1YDuemx3LG9ve4i7+xPo4Xxrx5Ttb3+CTfkTskrO1hxCdIaGDMjW1ttnfLD1gt/XPs3sG5ltGyplOCgQTGjJkiN12221+04XYE9xj9957rwt4UlJy+xuS8TDkxkR4iQXBG5Xffvstf056yJCh7m0Yu5mb7fjxMo+7gQcfHpDE1tiwYaPnRRAgYn3gGYeXHmnOmjXLXauBY7CEcQsnpggec3xSjoceesg/ceEmL9zGT57kzek9bdCgwR7ECC9J4mNgLZMmcTGoC9/vv/8Bq6god+cZvP/uvfcerwtW/zPPPOMu1rfffod7ECI2jz/+mEe1w9vxK1/5sr/9mkBRJSV73PmGmB94RuI5iPch5+AAg8WPcwhBjIghkpaWajfeeJPNmjXz1PZJ3i4EisIx5dZbb7MXXnjeY4GQFgGxcnPzbPz48d7GtNFtt93uwZnwlnzppRfdrZv2Y9DC0xCXdMpF3rj4E8eD4Ep4XjJrwOMQb0U8GwsLe1hZ2QmPi4JDEgGuuGb33XefLVgw3xoaGn3wwtWeutOmId7KXXfhUv+qv5Edb0q8NvFKxVty//79HoMEr9Sbb75Zz5aLTpHYFrWZrTlea1+cV2L/tPaotXQwcrBIEVLEG7HAVfuhhz7rbssEF+JmffDBB13E7r//fg8SVFZ23AWbyHmIZ3l5uT344ENuvSKExMs4cuSo3+y4c3NOc3OTR3cjSh4uyX369LaePQtt3bq1Xg7SP3681B599FFPj1ga3OwIBQLzzjtvW25ujrtqU+bAiRNlLpJPPPGE7d6926PhISQMGNDY2GD79u31MjH43HXX3T5g4Eb+5JNPeaAlglDFQrtSljb72te+bu+9t8Ij31GuNWvWekwR0iR2RigvcTloNwSfejLIBUuVgWHkyBEuXIgoLuWkgYCSBoMF7UucDuJpsJ3BC4hwh6DiDs8x5NWrV5HddNPNtnnzZj9m+/ZtHjXwkUcecRf5YcOG2wMPPGDLli31tIhNgriSL274BMWijLjCc42nT59m99xzrw8C7CcvrjmWMK74tCnpVVdX+WBMVEWuJ4MF7cIxDILkT9sx2NMnSAP3cvJicIqPuyLE+UhoixrbeUBuuj0yrLs9NbLQUjo4zhHukpsbYeDGQ6x5rT2xI4iah6WM6zdBg2Iu4Lg3sz58xAWOmBIIDfs4FlElkBLCgrVI+kSiO3LksO3du8+FAw++OXPmeChUbnggX2JNcD7WIpYiITexDhEPLEsi5Y0bN94DBgUQPY6nzEAUP6LLvf766/bkk0+eUVfKSohWoLwsecSLPvUgxgaCiwAT64IQpAQpwqL/zW9+bVdffbWL0sSJE2zt2nW2bNkyF0SsVmJoUA9mHggcsw6sZAY2AlshiuRLcKTgTk+sDsSN9LDUc3Jy3dqmfbDEY67nbR6vg3NYUgECIxHnBHGnrAhriP1BcCyOp11oN2YfXAssceKpbNq02cWcY2Iu8BUu9mwnDQY6zuE60ZZ8Yjlz7TiHeCcIMTMvysSMizIzUFN32oOgVMSN4bcBZj2xeCZCnJuE9kxsbWuzisYWK0iPxXzoCBYPNxzB/K+8crQHaMLiZRlhwoSJHoiHm46gPdzo/I14lpdXeLQ4pt2scSKgWMHz5y+wvLxcv5m5cRFDRIoble8cc/31N3gIz/Lyky7oWGLAkkUsPvZm+/rX/9JFEJhCY31TLkSAKT1xKRBC0qZM/ChImFHAauQ7wZ4YfAjryYDB4MKSAlNzBI2pPlHoEFAsVQaOBQsWeLkJABWLY93s31l/JVAUSw+IGGFgWTYhSBKCSRmGDh3iSzWs1/70p0+7qBM2llkCbUyEPerKchBiv2vXTg9yRexpRJoZB2VjYKNue/eWeGQ9yk5gKyIQIsj8zeDFzIDyUQbqUl/fYNOnT3fLnSUl9pEX14/ykebAgQN8ACYoFunF4ngv9mUYBiiEl5CwzGYoB205duwY27hxk6fJkk9ycpIPYFjSiDGxr6k3waKoF2KN8DPAIej8HYJ0CXEuElqoO8PFNk9YMw4vAoAQwOfD0DGQEEsFt9xy6zmD08cff6F0P0qZznbuhfIOgaiIA41ly0DCunFX5H2uMnTm+I5xUi71LaEYKOJCSKi7mHiB+LjS/rjSv1R8WuohxKVCQi2EEBEnoZ/6EEKIywEJtRBCRBwJtRBCRBwJtRBCRBwJtRBCRBwJtRBCRBwJtRBCRBwJtRBCRBwJtRBCRBwJtRBCRBwJtRBCRBwJtRBCRBwJtRBCRBwJtRBCRBwJtRBCRBwJtRBCRBwJtRBCRBwJtRBCRBwJtRBCRBwJtRBCRBwJtRBCRBwJtRBCRBwJtRBCRBwJtRBCRBwJtRBCRBwJtRBCRBwJtRBCRBwJtRBCRBwJtRBCRBwJtRBCRBwJtZm1tbW1/zvfvnMdc67jL5T2pYB8m1u7Jn/SKKlssPqW1k+8LEIkEgkt1AhGXXOr/XbHSfs/Hxyx1rPox9byBvvzxfvb/721v/K8QrO3utH+eukBW1dW137cr7afsA0n6uxSQJ7H6pqs8ZSYbq9osL9YvM/qWj66ODa1ttlfLj1gOysaPtT5R+ua7U8W7LUTDS3W0tpmR2qbrFWiLcQFSWihbmhps/++8pC9UlJu8w5WnVWoS6oa7ER9s31ueHf7/PDuNqZH1nnT3F/daM/tPGn/8/3DVn9KHBcdrraSqsZ2IUWcEL14a5tt8RZn/N/hO+IWb5H6tg5pNbeZ/dWSA7bhRL3/3Sc7zb40ooelJye1pxNfjvjyxO+Ppy3kYWb1za2eZ9hOec5lJYfyhf356Sn25KhCy01LttL6Zntq/l6raGw5ow06lqPtAmUTIhFItQQmJTnJvj62lx2ubbL/uerwOY8rzkmz6/rkWnKSWZKZJSXx/7mZUpRtjS1tNmdvhT08tKB9O0KzvqzOfrCx1MobW2xIXrr9pwm9rUdGqv3TmiOWl57iov6/p/a1l3aftKKsNJt7oMp6Z6fa48O62y+3nXBh+/8m9bGxPTJt2dEa++W2MitvaLHR3TPtP0/obb/fedKWHqm26lUt9sCQArt7QDd7e3+VDc7LsO+tP2Z/d3Wx5aQm25IjNba+rNZu7ptnP9p03Errm6w4O83+x+RiK8w83S2wzH+8+bgtOFRtV+RnWGVTi29HgF/ZU26vlFQY+slAdtfAbpZ8qm2o6/ultfajTaVW39xm9w3Ot7sHdrM391XauB5Z9q21R211aa19ffF++7PRPa1XVpofy2yAcvz3ScVWkJFiP9taZksOV1thZoqXrUdc2YRIFBLaok5LTrIh3TLaxeVsYIkuPVJjX5i7x76+ZL9tLY9ZqucjKzXZvjG+yH64KSbIAUT2b5YdsDsHdLN/nd7fGlvb7J/XHHXRm3uwyj4orbX/MamPFWWl2rsHq2zt8Tr7uynFVlLZaH+76rD96ZWFNqlnln1n/TG3bin1X48rsn+Z1t9Fe1Vprd0zMN+G5WfYU6MK7f5B+VbV1GJLjlS7FctyzJrjdX7ub3ecsIKMVGtua7MnR/Ww70wfYCfqW2z23ooz6rL4cLW9tLvc/v7qYruqMMsO1TT5dgac/7v5uP23SX3sb64qsm+uPeIDXqCyqdX+3xUH7YHBBfYPU4stMyXZl5kWHa7y/V+4oocNyE2z/3hVb5vQM9ut7q+MiJXjZEOLvbq33A7WNNlvtpfZ/7q62P5sdC/LSDn/ACnEp5WEFurOMKNPrs28c6h9a1o/G5CTbt9YesCXTC7EjcW51j8n3X6344S5MprZrspGX175zKB8G5Cbbk9c0cNWlta69c2g8edjetn4wizLTUuxtKQke3JkDxtZkGFTi7Lt1n55NrlXtt3YN8+O1jUZRRjbI8t2VTbYqyXlVt3U6ks0iDwWM+n3zk47JedmWSnJbtFybGlds209WW839821K/Iz7WR9i83aU24nG5vtWF1zex0YkFYcrbFb+uW5xf7ZoQU2siDT9y0/UmO1Ta1u+TNzqGxstcOnRBz2VjW6+DIoDc/PtIeG5FvKqQGRmUnfnDQX76Hd0q1beooNz89wgfZyNLR4ObpnpLgF/W8bjllDa6sPgEIkIur5ZyF+7RgrDtEblJdhDw4p8B/EeOqh/ZhzNGxaSpL9x6uK/IdKhBV8nTUuD0Q7leWUU0sq6SlJZyyrYOnzNwKHMcl3RA6DHiv8v6w4aMuP1rhg98xMPWdZnCRza/u9ozX+g+iIggzrnZVm311/zJ7fddIHBAaW9lHlFB2f7wg1aLU2G5ibbrf37+Zi/LObBp2xfh/axv914kkZxPi59nIwwJjlpSXbz28eZJN6Zdt/WLjPl1KESEQSWqgRzt2VDf6DYU1zqy9r1DS12J6qRl9DxXJmLXdDWZ3tqmiwX20v8+k/Fi+iwdrt+UQI63h6nxxbeLja/2aZheWO1/ZWeB6/2Fbm1irLKxcLPyzuqGywa4tyrCg71crqY5YwOo+luvlEvR2s4QfM0+Ub1i3Demal2rfXHbUHBxf4sdvK621Czywb2i2jfUAJMDBM6ZXtP7RuPllvz+482f7Ex9SiHNtf02ipSUm+ptzQ0urpBRjcKOPskgrbcrLefr39hC+zBKgzbUG6ZQ0ttq28wduWcgSr/ki75Z9n/XLSz7D2hUgkEvqXGYQCseRJDX7Y+/7GY/ZX44rcakUkWk+tKz+95bg/7TCqINP+6dp+bt2y1MCTC649cQLVKzPNpvfOif3oaGZ/M77Ij+XHvMKMFPvO9P72k82lvmTA2uxfjOnllvP1xbk+1Q+jZ/zfo7pnepmAHx7Zhxj/lwm97RfbTvgPbfcOyneLmHP/fExPLzPCyHLHTX1z3SrHev+TUT3t1b0VNr1PrpePtfTvbThmG0/U+zIPPxjGc3O/PNt0ot7+1/uHXbT/w+heVpCe6lYv6+P/uv6oJVmS/7g5sWd2+3mU/VvT+tu/by61V0rMl01YkgkDU2Zqsj01stAt6b8cW2R/Pb6XW/cI97Q+OT6oMMYwOJQ1NPuyD/UQIhFJakvgZ57OV/V4/UWwOTT2WxbLEaen9h2fAolPk+1n+5stLHuE5YyOaZ3t747nt5et7fTSSfwx8dsvlHZrh+lVx6daYo/Zhfr/YT6kF+rS8TxPuy22Lh3fruH8+HQ7HhuOb+3Q9kIkGgkt1EIIcTmQ0GvUQghxOSChFkKIiCOhFkKIiCOhFkKIiCOhFkKIiCOhFkKIiCOhFkKIiCOhFkKIiCOhFkKIiCOhFkKIiCOhFkKIiCOhFkKIiCOhFkKIiCOhFkKIiCOhFkKIiCOhFkKIiCOhFkKIiCOhFkKIiCOhFkKIiCOhFkKIiCOhFkKIiCOhFkKIiJP6SRcgUWlra2v/npSU1GXHdhUfd55dmX5I61K3zaXKT4iEtqiPHDli8+bNs6VLl1p9ff1F3ah1dXXW0tJy1v2k1dzcbEePHrW9e/e2/92RBQvmezpno6mpydatW+d58Z1yni2Ni6G1tdXWrl3r6dXW1p4hlh1h/6JFC+3jhPpfTLufixMnTtiKFSvs44S2qq6utsbGRtu0aZOtWrXqvO0XOH78uF876sm15jv/jh075ucfOHDAFi1a9JGvrfh0k9AW9caNG1yQhgwZ7H8fPHjQysvLbcCAAZaXl+fbuDkR2379+ll6errt2bPHv8+c+bKNHj3GRo8ebfv27bWePXtZ7969XQxfeOEFGzRokA0fPtzF/KWXXrT+/QfY2LFjLSsry1pauCmTbPfuPXbVVRM8Xxg6dKhVVlb635TBrM3T27Vrl2G8cWNzg3NMZmamlwOoQ0nJHhswYKClpaXZ7t27raioyHr27Nm+P5QbDh065OX/wheesMbGBquoqPCyZmRktosH+ezYsdOmT5/h6WVnZ1v//v3dikR0du/eZXl53XwbA15FBe020Pd169bN0+STsufn53tZGLgYJAoLC23//v22ceMmmzx5irfx4cOHbciQIZabm9suvkeOHLahQ4e17x88eLClpqZaWVmZVVVVeZm5JtRv//59Nm7cOG+HXr16WVFRb0+HfChTU1OjjRp1pZ9LfQYNGujlpy2oc0ZGRvugkZyc3H49duzY4fmT1w9/+EO78sorvZxc1507d3qZuF7UsaSkxM+lT3AdSPeNN163559/wfsT/eMXv/i5XX/9DTZ79mv2j//4TXv++eds8eIlfm2GDRt2ie8AcbmQ0BZ1a2ub31wHDhx0a+fpp3/iN/bMmTNP7W+15557zkXr2Wd/b3PmzLEDB/ZbbW2Ni2VNTY0dO3bUv//+9793EeLmRERi4nLIhSr8vXLle/73unXrbcuWLZ4HIlNaWmpz577jokA6DBYI08qVq05Zbyv9/Hfeeduef/5527Ztq9/g5A9z5861ffv2+d8vvfSSn/vcc8+68FCeF1543srKjrugUAbqWl5e4Z8HDx6yLVu22uLFiz0tyjJr1iwXa8qG9bd161Z77bXXXGiDiHI+6TKIIfoIIOVfsGCB73/77bddpBggXn75Jf/8yU9+7HWjjqSFSCLo1IXP3/72t97mlJv2Jg/O4xrE9v/Gdu7c4XVbv36dLV++/IzryXWqqDh9LeCnP/2pH//tb3/bNm7caN/85j/a5s2b7X//7//js4u//du/tc2bN9mPfvRDH1D/7u/+p7388sv2zW9+09tgzZo1NmfObN926NBB27Nnt9eTNJ5++mm/jj/4wfddtEnnpz992g0AaGho8Dwff/xxn51Qt7S0dB9E+U6dDh06bHfeeafP6jpjoYvEJKGFGutw/PjxNmHCBP+OFXv99de3CxzW5t69JS5Y/fr1t4kTJ7qQHztWasXFfW3kyBFWWooAxgSH6SsWX9++xTZy5EhLSUm15OSkU8fG/uYGxaJua2v1MiDAWKd8IpJYZDNmzLCBAwe61Ub+EyZMtGnTprsAY0lfd931VljYs90CnDhxgg82iDVpxcobs54pU2VllVvGWIMtLa1u3fbr19cFIwwkCChQBqxN2gErGlGqr6+z3r2LLCUlxY9B9MmL8xD/vn37efrdu3f3ulFH8uVcBoPi4mJv3xEjRri1iXV7ww032ODBg1zgmY1Mnz7N2xuBra6usszMLJs+fbrPQLDA+U6aDQ2NbjlPmXK1lZefPON6YtkH8T+9LNVmDz74kFu+lJk0HnvsMS8Px3EN77//AUtKSraHH37YZz58FhQUnLK+d7ng8m/IkKGnypRpKSnJ3h8YGHNycvy6M+hzHIMF/Ye2xXpnYH3//fd9QKdulZUV9o1v/I0PgLQf4r9kyWI/V2ItzkZCCzXTZiyiJUuWuOghDtzA3Ih8pqdn2IgRI/1YptOIWX5+ga87FhTk24oV7/lUGwFDUMKPSwUF3W3lypUuFlhQ3bsX+N89evSwZcuW2urVqy01Nc1FF7FAzJh6M30GLHpEjnKQP9bWm2++aWPHjvNtiDnnMgjA8eNlLiwnT560MWPGulAi5Ey/ERCEbtasmbZ69ft+PvVmYNiwYYOVlh7zv8kfEFXEhfyozlVXjfd6UG8ECQ4fPuKflIFBicHklVdmudAyYGD50ybjx1/l5yLgnEv7kldubp4fz6DAcgCW8CuvvOp5s79bt3wfHCgzgwEDw6uvvuIDC+lwXUifT+A75T948MAfXAuOpx1ycnKtT5/evrTy3e9+x69ncXGf9iWu3FzENs3y8nK9zfikT1A26smgRV1ff/11Hyyoy7XXXmubNm20adOmed+grOSTnp7maWKxf+5zn7evfe3rfh2ZHbCUwzaWxtatW2tf/eqf2l/91V/7MtEHH3xgixcvkliLPyCpLYGHcCwYrBxuakSSH4q4yRFObkzAwsOy4gZHBKqqKl0Ew9SVG51PjucY0uIcblxuPkBY+RvBQkwRFW5+juM72xAo0ohZwJUu6pSHbZyL4LGNslHG2MASG1DYxjlhTZqBhGNC/uTDcgT5kz5pYtVh4fM3dcnOznLxoTuQFttCOak/deB8PkN65IEQshRUX9/gAwLnQhB/zkUoKQvtzTnUK7QZ/yg/Ikf5g9UetpEm54XvQFsgppQjtubf4mlSNtbKs7Ji1wI4L769OIf2Ji3yIh+uIdc1OzvHl4+oU/jk2DAQkCfXIog76bGfv8mbduc46hQGQ/6m/mGZihkHAxHQVuRB+nwP69phQBQikNBCLYQQlwMJvfQhhBCXAxJqIYSIOBJqIYSIOBJqIYSIOBJqIYSIOBJqIYSIOBJqIYSIOBJqIYSIOBJqIYSIOBJqIYSIOBJqIYSIOBJqIYSIOBJqIYSIOBJqIYSIOBJqIYSIOBJqIYSIOBJqIYSIOBJq8ZHRS4KE+HhJaKHmbd379u319/p1ldjwrsFNmzbFvQX7o0G5SJNP3jPIOwCjAmXinYG8DPdSijVtu3HjRn8/YnhH4/mgbLxIl3cvftx82GvEebz89uOAN51T/67og2eDt8Hv3bv3vOeTP/ca76jkbfEhTdE5Ui2Bee+992zXrp1WU1Nr99xzjw0dOvTUm6R5I3WqCzkvs+ZN4uGt1sCLTMNLU/nOi1XZzzm80HThwgU2bNgwf6EpLy7lZafsp3MiMmznpae8AJWXrfJJGnyyjzTDm7Q556233vQ3eiM0HMM+8uEFruznBbX8C3+HsvGyVL5nZKT7G7tJj/y5QciD7+H82NvXs7zObOeFraHO/E061IUXunIOdeW8/fv3+9u9eXksdQgvegWOo5xpaamWkZHZngbtQV3Cy4RJK9SV9qAMpBNe/suxoewcQ714W/cdd9x56g3jOZ4PnxwTXlgbrhNtdujQwfY3rYf9IT/+8TftEq5/7EXGVV5fzgfKQRocz37+hfbjupEGL/2lTNSR88NLdjmH40J92M5b23mjOW+GR8gY4Hk7OYMP+XIsZaVclKljG/HC4JBueDN9OCa+Ljt2bLe8vG7+5vX460kZqQNpUKb4NggvauYase3dd9+16dOn+wuWQ7tQPupdUrLX3wA/cOBA38f1jO+/69atsxUrllu/fv2sb99+nh7l5+311113/RnXgLrEXz/K2tLS4m0a2j/+XkwUElqo6RS8+To5+aR3yDlzZltjY5PfQNdcc629++5c6927j7+hnONCpz158oSVl1d4Jx09erTNnz/Pbr/9DtuwYYN3/AMHYhbM6tWr7cCB/fbEE1/0N3jDK6/M8o6IwD355FO2cOFCe+CBB+zll1+2iRMn2urV73uHvP/+B6yoqMgFcPnyFVZbW2fFxcV+Uy9dusTfIE5eHLNmzQfW0tJqDz74gG3bts0qKir9zeBTplztgxHpTpgwwfPfvXuXHTx4yAelLVu2+Juzm5ub3KL7whee8BuK8t966612xRVX+I22YMECKy09Zrt27fayHjly2EaPHmNr1qyxpqZGFwCOOXTokOc1adIkP4/2WLVqlfXt29cKCgps+/Zt3mZ33nmXzZz5svXp08dvus9//gvtovDcc89ZZmaGFRR095uZtI8dO+qC1tBQbydPltvjjz/udWEmhChQDuo7fPgVLtiUg3YpKMh38evXr/8pYcu3uXPfsZycXBeaYcOG2rx5810EqC+DNm9Tp83HjBlj69evs8mTJ9uYMWM9P8rM4DBv3jw/HtGiD73wwguWnp7mg1FRUS9Pe+XKldbW1ur7r7hihC1atMj/RjAp43333WczZ870tuF6I4Jmbfbmm2/6m86xUh944EGbPfs1GzFipF1//fVe1/g26t69h1un9Emu35/8yZ+4+ALp0M/4R7+Jh+uOeNOWN910k7300ktel8GDB3ublZSUeJkwDl577TW79tprbfnyZd5OQ4YMsa1bt9qxY8fstttuszfffMOPHTZsuC1fvtx27txhVVXV9tBDD3ndYMmSJfbYY4/5taypqbbZs+fY0aNHva8PHTrMNm3a6FY/+axfv96vE/WlzzBje/fduXbbbbd7uz/yyCOWiCT00gc3BkJ19OgR70SISkpKsls03PzcPNxQra0tp6yfBhcEhDM5Ocm/c4NMnXqNDR8+zMUAEenbt9hvKoSUm45jAKsBkXzsscetR49Cv4mxQGJWTKWtWrXSjwlLCoBlMnLkCJsxY4Yff/x4mXdmyon1Rp7XXHONTZkyxfbu3WcrVsTKFqx5yh1vgTAQ1dXVnrLKavzGv+eee62wsPCUNZTm4osFGuA71lv//v38ZiXPmBVZbdXVNV5+rE6EaN++fe3nYUndddddduedd3qZ77vvfrv66qm2efNmrxc3HW0TpsAMbggI5ea6TJky2Z555hkbN2681wOBpy2YQgOCiujx9yOPPOoDBLMHLLE9e3bbunXr7TOfuc9FlTKXlOzxAffRRx+1/fv3edmvuuoqu+GG612kEdeYFV55ypqtt6Sk07fIoEGDbdmyZVZZWeGDN1B+hCY5OcUHDgZUBvEPPljtafGdvBkoGThHjRrleTKY0AaUpbq6yutCmyJ0tAuWJyJMe3OtArE22uFtxHfa5d5777Xu3Qs8ndDPVq58z/sB27jG7T2+rc2vFWWjDRByBqzPfvazPpCGfVjJlImBFzEfMWKE3w8MQuynT3B9p02b7oMX2/ibwYVzMBhCfpSHekBrK7OsKk9v4sRJVlzcx/dhNR84cMD7Gdc0zMowcE6eLPd+PWjQQEtUElqouQnpLNdeO82tPywPbsabbrrZ+vXr6+ugH3zwgVtKW7duc7HhHPb17z/AOzBTSzo3Swv19XV+PALKuQg71luAm4sOiAWMpR1bmqh0ywIrBAukZ89eXp7+/fu3n4foIzwIIRZ1UVFvt2BuvPHGUwNCanvapDFgAGW72Ts9FiEWeFhfpLysJwYrnHNi/1LcUkbMsK7MTos7N+yePSVusSIu+/cf8Ho2N7f4IMBgtHHjBh/s4geFwsIetnbtGlu7dq0LG9+5gZkCx5Y7YnkHmG6TBmKGKNCGN998s59PGRA/rL3Cwp6eT1huou4MrFxDrDqWYqgP4kU9sYwZlPPzC9xyI73c3DxfkuEfx5JOcXFfX3oIbTdhwlXedrG+kmRjx471ZS0s3FBu2gPrfcSIK9wKZDvWNbOFIUOGeh/hOsfKmdz+CQgxs64gnlxH+hHl41r06NHdJk+e4gMB4nW6jXq2t1FYgoktw5zuZ9Rl4MBYXcKSSADrHms8LOlwTSgH2xYuXOT1iV0faz8mNvjtsSVLFlu3bnk+kHF9sc43bFjvdcAa5zowkGCshHYbPfpKn60yELKPbaRLHXfv3uP5c02Bfo2wM/ML+Y4ePdoWLJhvV145OiGXPSCpLYF/sg9TZzoN35k2bt++3cWC9TamzXRIpnt85yahE3Mj0EGZmvE38B0RYErHDdarV5Ft3brFhZopJTc0TY2lSId/770V9qUvfdlFHUuW8xkosDY5bsyY0X7T8p1jWCpBvLnpY8K526f0lJlt4cceOjZTU0SKMlAfxKd3797t67zUBYEKVj31LSs77lNOLDqsagQfa4bjf/KTH/tSECJFmREY8kHQyZfjuLGwlsgntAnlRMARAOpGvqw9Dh8+3C1K6oMVxSfpUX7+Zr1z8OAhfj7lYNuiRQt9uYZBiiUZ2prrEdYxEQB+F8DCZTbRs2eh54vY9+5d5L8zUF4swdLS43bllVe2/2YQ1j+5rogGVjd1oi2YmvN3aLvvfe979tRTT7UvZZHG4cOHffCjHgxGAwYM9O87d+709uDYcJuRDm2Glfvss7+36dNneP9gAGWGwjXn+tFeiD0iTZshYB3biIEAUSMPrGtmRfRXjmHwxEJmG+1AP2ewPP3D3j7vc5QF8R03bpwvZ9F+1Ic+HPoH7cZSB0tQDAD0X87FIOA+wHhAaOmLXGP2cY3DYEYa1AnrnWtHf6d+DKzcZ1wTBm3qEWaJLD2FQWLXrl2+REi7S6jFJYObBStv5MhRbu1EvaybN2/yZRUEb8KEiS5ol7oMCBaiGdZgPwlYZli/foPPUs4mGAwsixcvdksX4T0fiBczNJYJPknx4fcWBi6WIqLK3r17vc/FzzITjYS2qIUQ4nIgodeohRDickBCLYQQEUdCLYQQEUdCLYQQESehPRPjH5nq+Jvq2baF7RebB/9CevHPDZ+tLPF5xJevM/W40LGdTe9C53yYdM5GU2ubtbS2WUZKktW1tFlGcpIlJ53/erA16SNeD9xrwlXoqrbtbN4fNa3m1jb/l5kqGyuRSOirjScgz/MGeOgfhwo8FAPz5r3rz1jj4PFhbkyeUZ09e7Y/H82zpOcCT0QcEeLhWVieH70QPHf6zjvvXPA4nqmmLgGeHeaRsvPBc8U8Dx7P7spGm3fwdOChqqYWa2xpdeGtbGzpdNvMP1hlHxyvtdY2s59sLrUFh6psw4m62H4zW3Co2jaeqLclR2LOHvDCrpN2rK7ZFh+p6VQ+HfN8eU+5/W7HCZt/qNrqWy78wFMoB8d2tm7ngnqSbyeyPaPMJxua2wf8+YeqbPXx2o9UDnH5kdAW9dGjx6yyssrFCweImHt0jW3bttW9x0aNutLjXuAIgMNAcKxA4Dge5wCe8eTZXpwTcArAawsvQwQQR4b58xd43AMcHHDI4HlgHEM4nzRxOCBN3NjxVuQ8BB0ngxBMCMgboeVZXjy0yAunAZxPYt6C+708OGlQFpxhOAfnBZ4/xUEiWHFEu8MphE9EHhdrnlMlHbZTJuqFgwkDCB6RuE3joYlDRWVTi605XmepyUk2oTDbfrDxmI0oyLQeGSk292CV/dW4Ihc1BHVQbrp1z0yxNaW1Njw/03LTkt1qzkpJtlXHauwvxvayAzWNlmRJVtXUavtr6qyysdUm98q2tOQkq2iM5YXIsW1nRYNd1yfX0pPNGlrabHVpjWWmJtlVhdl2qLbJ9+enp1jPzBQ7Uttsg/PSbX91ozW2ttnw/Ax7raTCy1d/amBZd7TGyzMiP8Oqm1qtT3aa7atutIL0FFtXVmcjCzK8HJR1zr5K+8b4IjtZ3+x1m1KU7WXdUVHvZctNS7GaphYrb2yxHhmpVlrf7LOEI3VNVtvUapN6ZVs6oQeaW21PVYMlW5KN7ZFptc1tVlLV4MdP6pntM4w1x2stOzXFemWl2v96/7D95dheNqog01Yeq7E/Hx0LriQSh4QW6paWZhdIRAwPLyzqqVOvdq83vKSwhLGpYrETVrrY4oGHRxzCiAiOHDnSrW1ifSCkd999j8cy2LRps3tgBZdi3MTxACQWBMcjwpxLBDGsZoL5EHiIIDx4CBIdjkBPAfLGjRhPO1y/mQ3gJfjGG2/Y3Xff3W6BMwg8//wL9vWvf923vfrqKx5b5I033vQ6ES/hxRdf9JgMwZsRt3ECOTGwMEjgVcaA8NJLL3qcDaBuiD+DAzN4hKZ7RoodqG6y5jgLkWl5U0ub/dOaI/bYsO72vQ3HXHzqWlptUF66bS9v9Gl7XlqK5aQluyC+vrfSpvfJsQM1TVZS2WA1Ta12rK7JhfKmvnm2uzKW146KWJjS2pZWt+i3nmywrNQk21fdZBWNrbbwUJVdVZhlK47W+PJIQUaKC29lU6stP1LtQpyC+3JKkr2+r8qKs9NcNOcdrLQ7+ndzy/nhoQW28FC1nWhotqlFOS727xyotBuKc13Yqfvx+mbbWdngAwNlQ6TZB5SFGcDt/fLs1b0V1isztX2gSElOsrcPVFq39BT7tw2l9o1xMQ9OBqofbCz1NiC93NQUHwwZMO4c0M0aWmLlPljbZNkpyd4WIrFI6KUPYitMmjTZLebjx0vdcuRGxNolxgIRzALsw+JFuIgPgjstwZtYKiEYEPEiCM6Eu+/GjZvcMsWiRZwRuBBPmiUQrHYsWTh2rNRFnoA9CHoIhoMXHhZ3gPRwz+VcXI2DyBL5Lrb22eauvQwgDBTBNRoXXSKpXXHF8FPrvOblocyILvszM7NOhZNs9k9mFJSL2UZs8m/eTljqgGF+TVGOzSjO9WWPfjlpdmX3TBdiLNjCzBQXwVv65bk1enVRtpU3tNiq0lq3Kkd3z7SlR6ptRp9cFzBEb2yPLO+M1xXn2ow+OS6Cp7TPBfPGvnl2tPZ0cCL2YYUi5FOLsl0My+qbXcwRVazsewZ2s5Qkc2sYsSa5vjlpNiQv3c/HgsUir2ps9cGDss7cU27X9M5x6/7Gvrk2rjDLjy3OSfPZQbf0ZHvvWI2dbGjxOn1mUL4tP1pje6safb0d+GTwavHfJsxu6Jtr4wuz/BwOYaWdMmNN+99t5gPMDcV5PhPZWl5v9c2tNrRbhrdtv5x0G9ot3ZYerrbpfU7HjhGJQ0ILNZHW5s+f73E3WOYgLgeWLvELZs2a5QIeYhOzj7+xfhG+q66a4FHmQgS+WJjLdA/FiZCmpsZiN+fmxuIIkw4CTLwDBDTEMcCK3bBho68xs6yAezKiHOJOB5dpzo1FxEt1YSV4E2E3WZpA0CkfcRKI9BdiF/MPgcUKZ4AJMarz8mKxhonngCs75WYdmoA8Ie4J4s/MgdgT/CN6GTMNQNCwhvlkKQOr9a39lb4Usquy0XZVNrj4/nBTqRVmxpYAOB4RwtpddazWlwvG9Miy7eUNNiA3zTJTklwsSeel3eU+ELCEwTLA8iM19sutZTa1d45bo6lJSf55dVGO/Wxrma91T+mV7aKPJX2wpsmtUspT09zqosuSQ1ZKkqeZdOp8LGyWHcgjMyXZLVrWxMcw6OSm2482ltqCg1V+Tre0FNtb3WibTtRbWX2L3zjUCYHOTk32PH61vcxyUpN9sHh+10lPk30ZxBdPTfY65qcne1lYJuFHTc6h3LQjvw+ynQGiurnVUpPM82X7OweqbHtFvY3rEYvzLBKLhHYhZzmBf0GUWHsOMX/5RwQ1PkOoSYSU5Yfw9MbPfvaMCytC+OUvf8WPCwHn2c/xWKghWDz7OZ882MYnaSHsPMvAfv5heRM0KSsruz3YfOyFAJVuuYcA+OEFBPyjnORN+gwQ8UHksd4RfOpAueLTZFZA3Ql9StohwBMDDmvSIa4H20KUttWltT4tRwhZ6uAT65IpOWu2rPnO3lthdwzo5iJHDytvbPZ121aCAtU3+7LGPQPz/cexoXnpNigvw5cPWDsmzYKMVLcqGQzYxno0a+D8qIew8jfCd6Khxf/mBzpEnjxfLSm3h4d092WR5CSOaba0pNhAQNqIZl1zq5+Hlcsn26jT1pP1dvfAfD+OfSyfhKcsKhpaPE3y5qYhf/bVnypbbUtbTLRP/ejIEguCThnCTUYdM06dl5GS7Ms8sfK3eV0bT23nB8TY8k2ql/VIbZNtOVlv9w7K97X2kQXRjhEjupaEFuqPQohCxuug+vQpbn8jRiLUG8sSETsfLafWZM+dUOw5O3pfVzUbYsdSAhb3h0mTQYQfNSN7GU+1WZ+sNF+KEYmDhFoIISJOQq9RCyHE5YCEWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIk5qVyXU1tbmn62trf6ZnBwbA5KSkroqCyGESEi6TKgbGxvtwIED/olop6Wl2cCBAy0jI6OrshBCiISky4Q6NTXVunXr5gKNWPPJNiGEEB+NLlPSlpYWO3bsmNXV1Vnv3r2tR48elpKS0lXJCyFEwtJlPyaGtWjWpquqqqympqarkhZCiISmyyxqBDozM7P9R0QhhBARE2os6vT0dBfqpqam9qdAhBBCRGjpo3v37i7StbW1LtpCCCEiZFHz/PT69et9+YMfEsPz1EIIISK09DFgwAC3qHnaQxa1EEJ0DV36y19OTo7l5eX5+vTx48e7MmkhhEhYulSosarDY3r6MVEIISK29NHc3OzPTuONmJ2dbVlZWb4MgoeiEEKIiKxR451YWVnp1jRi3b9//65KXgghEpakti5co+iYlCLnCSHER6dLoyax1LF9+3arr6/3yHlFRUVdmbwQQiQkXSrURM3Lzc11yxqxFkIIEaGnPlifxsmFZ6hDeFOEWwghRIRcyHnCA/dxngAJ34UQQkREqAnGhFWNQPft29fFOj8/v6uSF0KIhKXLhJp1aV671dDQYIcPH3YvRT31IYQQEVujrqio8DVqBFpxqYUQIoJLHydPnnSR1iu4hBAigkLNM9Q85YFVzUtu8UwUQggRoeeo+RGRONT8iMhjenwq1KkQQkTQhRzLGqFGuLUEIoQQEfNMrKursy1btvgPizyip6BMQggRsXjU/KDYr18/fx2XHs0TQoiILX3EJ4NlzRMgiLYQQogIvdy2pKTEn/xg6YMfFoUQQkTUom5P/NRruYQQQkQkKBOWNPGoy8rKrKqqqquSFkKIhKZLf0wkBjVLH+Xl5f5PCCFERONR895ErVELIUQEhbq0tNSf+OAtL7KohRAiYkKNu3h4hpofFsNbXoQQQkTMhZwYH1jXxKbWUx9CCPHR6TKzlxcGHDx4sP0xPd5AnpeX11XJCyFEwtJlSx8EYerTp48LNj8m6sW2QggR0XcmsjaNaFdXV3dV0kIIkdB06XPUvCyAJQ9+UOzevXtXJi2EEAlLlwp1WKfmCRDFohZCiAgKNU95BO9ElkGEEEJETKhZn8aarqmpOWuQJiGEEBdPl3qlYE3zYtuCggL3TMzPz+/K5IUQIiHpMosaCxonF9ap9+7da4WFhV2VtBBCJDRdJtQEZOKRPNapefqDl9wKIYSIkFDzlAfr01lZWe74onjUQggRwR8TeX4aq3rnzp2+Vi2EECJiQs1yB098sE4tz0QhhIigUEPfvn19+QOXciGEEB+dLlVT1qixpnn6g5cHCCGEiNhz1LzdhXjUvEAAsRZCCBFBz0ScXo4ePeqfQgghIibUxPfAomaNWkGZhBAioo/n4eyClyIOMEIIISIm1KxRs+SBZX3ixImuTFoIIRKWLhXq8ELbY8eO6X2JQggRxaBMCDTPT/fs2dOtayGEEBGzqHl2mvcl8oOiQpwKIUTEhJolDx7Pw7JGpPXUhxBCRMzhBYFGnMPbXfBS5CkQIYQQEYpHzYttEWe+61VcQgjRNSS1dZGini0ZlkOEEEJEZOmDN4/v27fPn6NmrZqXB3Tv3r2rkhdCiISly5Y+WJPu1auXeyb27t1bb3gRQogoPvWRl5fnj+bxclveRC6EECJiz1ETi5qnPnjTS2VlZVcmLYQQCUuXeibilRje8KLnqIUQImI/JmJNl5SUWG1trYs0PygKIYT46KR2ZUAmrOnjx4+7ZyJLIEIIISL4zkR+TMTxRY/mCSFEBD0TeYYasWatmpjUQgghIiTUYV2atWoe1auqquqqpIUQIqHp0l/8EGos6oEDB+pVXEIIEUWhZvmDHxP5LCoq6sqkhRAiYelSocaaxoWc56j1clshxKWgra3NSqoarayhuVPH56Wl2Ij82GsDE1Ko+TERgeY1XHrLixDiUvH3q4/Y73d27oXaNxbn2ux7hlva5aPTXSvUPEvN0x4Idb9+/boyaSGEOCsEWG5qbbOGls5FbObYhH6Omic/iJyHUOupDyGEiGCsD4IxnTx50sVaL7cVQoiILX2w5FFdXe1vIcd9HKuaHxeF+DgJbxb6qD8Mxb+h6OP8keli8+lM/eKPudj2OFt5PkybxqfDd8492/kd87vYF0wlXUY/AEbW4SUnJ8ff9ML3RG1QcWnBOOCRUG74jv8IEEZ/PNu+jv/gwIED7e/7/Dj+AT+yb9myxb/Hbz/XOfxAv3379vOmi1F07Ngx/75z506f2V5MmTZv3uyGVti2bdu2Trdb+FdeXm4VFRWeDmU4Wz58sp/8wt+wadMmb/eysjJbtWqVrV692q8Fn8S237Vrl61bt85OnOjcj4WfRrrMokaYaWw6Ii8NUPQ8camEmhsYoUKwhgwZYqWlpT6zO3z4sH/y5iF+M+nWrZsVFxfbkSNHLDc310UA5yz+5pHSkB7CwN+cx+vlmBkOHTrUduzY4dt5QQaDw8iRI72fI7zkPXjwYI/DjmgNHz7cj+/Zs6fPMLk3Ro8e7Z8I2qFDhzzfK664wstFWdiO/wHbeZUdXr7UAUFjP0JGnpSJMA0sL7KfY0mX7dSd7ZSPgYr26NGjh0e2JGgaxw8aNMiXKEmTOpIvZd6/f7+Xl3yoB/uoK8eRBnUl3VBG2o48SIvzeJgAkaet+UeeDBpXXnmlP7ZL3ggw+0iD75SdvKkjukH7cm3QEdJjGXXVqlVeb9opUenSWB90ci5IsKqFuBRwkx89etSfNEKsEBqEDHFhGyKOSCFi9E0EAisVkcRihf79+/txbOdYxIPjmCUiZAgQn4gZ6SBA8XkjfliSGCwMHGxDoBEa7gXEDuGNLzN5E8AMED7yDdbt7t27XegQfOoCiCfbEUcGBfIbMWKE76dM5DF27FgvK+mF9gDKwvYQ1XLjxo0uhuQRrF7KyeBCHRiY2McgRB6kzeCCsDL4cTzHcsyoUaPa68KARVtSNtqQ4GyUBdhGmyP469ev93pyfkfPZv5RBvJjYGhubvY0QpsnIl361AcXkAtDg3OzCPFxww2NUGGJZWZm+t8IHoKG8GDt0R8xILDQ2B6ChyG6CDeCgrCSDschCogD6fF3cOTifIQTaxCjhPMA4USk2I81Hpb+sAwRLgQnfoZJPogWn2GJkHuHPLAa+Rw2bJjvwwpF0EK6HEfebCdN6kAefEc8w+BC2WmTkAfWL4MAIgpYyhxDPpzLebRLGBSoD3lxDPv4TjuQNm0YhB0oAyJK/RkIGPBIP1yTeH0gXYw69lNPBoT4pdKwtk05EHXKm3aqHFyTzpKcZHZLv1x7eGiB5aV1qcx9IiS1Xexq/jmg8em4dCo6Lhdfz1KLjxtuXv7R/xAKhBfhCjc302w+w42OVcp3jkXEmW4jLiGoGGLEdkQmWHZBhOjfQSzJA4ud22fZsmVuVWKkkBb3ACLNZ8iHtBGneMcwLGyO4xjSC34IHM92YDtlCfUM4k++bMfKD0sCbKO+lIt2CO3BceS1cuVKmzFjhosgaWH5I7zUkXpxLsfyN/cweYXBjjxInzYhTerJueG3AI4lXepGe9Gu5Bmsbz45L+QR2jnUM+TLOaQTZCkMTPX19X5sWKI6Q3va2uzL8/bab3ecXsPOTUu2vxzby47WNdt7R2ts08nYzAlu7ptrb33mCktDzRNRqGl4OhqdkgsVRmchPq2EH/OC0EQV7kuEMsw6Pk20nkWoM1OS7P+5qrcVpKfYDzaW2t7qxstaqLv0x0QsCkZqPfUhEgX6PVP4qMM9eTZr9NNKQUaKNba02prjDTYwL/0Mob4c6TKhRqBZu2JKw/RlwIABPjUS4nLhbM8j+/ZTn8kf4tnkkN7Z8jh/GqdzPvvzyOfff6Gyne8cLNSw96M8j32hPNsf2ztVHwzccz3L3cYPnhdxDY7WNtuLu8stIyXJdlac/hH3cqXL1iZYe2JqxeI/Sx4sgwhxOXGwpsleLalwoSqtb7ZZJRUuDu+X1trh2tPPJpc3tFhV4+nnjjv+O1zbbDsrGzz2xKrS2jP2fXC8zupaOvN8c+zYEw0tVlbffNb9a47XWW3z6ee+A+dO12z18doz8q9vabXSujOfu151rNbWn6iz9WWxtfBO/TOz947VGmE0Oj47vfxojW0pr2/f3tLaZodrmqy8scXXjmfvrbA5+yrOyI9rQUS88Pfe6kZfa+4s5L6nqtG2ljdY8+UX2uPjs6hZ9GedmkeAWA9L5GcexeXJgZome2l3uV1dlG0LD1W7yI7vkWUflNZaalKObTpR71bfrsoGa2xpszsGdLP3j9XaqO6ZVtXUYqV1zXb/4Hw7XNtox+qarVdmmotebVOr7atutCu7Z1pdc6utOlZjY3tk2eaTdZaenGw7KuptWu9cF1HWVGuaW61vdpo1tLTawkNVPlA8ObLQyzOsW4YHFTpU02iFmalW3pjqg0taSpLdPyjfUpOTbP7BKs9/Rp9cW3Kk2o//7NACy01NsdWltZaSlGSbTtRZRkqy9cxMdaHkh7f3jtVY94xUa2xt83JgvM4/VG3byutteu8crwP7JvbMtmVHqj2vOwd0szf3VVr/3HSrb261AzWNtvhwtdd1Us9sz5u2zE5Ntv456TZzT7n/0PfW/kr72phePhC9vrfSHhve3etNXbdXNNiogky3ht/YV+ltO6Ig08oaEvf1fl1mUfOLLM4ErNcVFhb6w/JCXE4wqZ7aO9ve3F/p1l6/nDRrtTbLS0+xeQerXFA5ZkhehovVqyXl1tDa6sK07nidTemVben+AxVLJ6QYM+WwTq/pneOW5bbyBqtparW391e68L+4+6SLH8KHIE7qlW27Khp8QODvoqw0m1CY5UJH2nMPVLrlOblXtluciPLA3HQfALAgTza0uPAer2+290tr3Brvlp5iuytPT/+xZkmPYxDNqwqzbE1ZnQ9UK4/VuOBiqW89WW8rjtb4DONQbZMtPVLjywhrjte6SFc1tdrcA1Uu9uMLs2zTyTq3XtnHdthdGVsbRsD3Vzd6eW4szrPR3TO9XY/UNtnI7pne1tR3waFqu3dgN8//UE2Tp/UBM4v6zj+a92mky4SatSSeL8UTisfy5PAiLjcQRyy5tcfrXHgykpNs2ZGYcKUkmU0ozHaxQjz2VjW6iOanp9gt/fIsLz3ZumfEngfuk51qG0/W2TsHqqx/bpqLYo+MFBdazh1XmGULDlW5KBdnp7n1jJAXZqRaTmqyTeiZ5QMD67EIKUspDAYsuXA+x/TISPWnFvjcWl5vZfUtXhas0F5ZaW7RjizIdKub7WFlJDUpVga283xxZkqyW9+5qckeUJ+65KQle31pD5ZF6lvarHdWmtdhalGO9ctJd3HmfAaBHRUNLuqUh+WJ6qZWzwcQ9alF2T6IrSurs6KsVE+fwYr6cE5WSpK3Dd/JlyWUmqYWq2xs8YEjIzn2bDX7EpUuezxPiMudsN5b19zm4lbZFBMSrD4EsaY5JpSI1IHqJivOSXPB7pWZajyJmp+WYinJsR8isUARrKHdMnw9G0GraGxx8UMQfWkkK9UtYaxNlgWaTi057Klq8PyDKO+vbrLCzNg5PM1AmUiDJQH276uOWcjF2bGVTNbXT9S32IDcNE+TdFjuQJRPNjT7MgkWGpY852NJ98lK8zIj9GHZAyH91fYyH5CoB29FYU27T3aa59Pc2mZZnF/d5AMVeXA+ljBLKD0zU3xpgzKzLk39KQvp769p8rwR9Oa2Nhd92h949pkBLOlUXVKTkqxHZmywwQrvzON55+NyfDxPQi2EOCsMOEdqm62updUG5KZHVthaE0Cou/QNL0KITw8sNzBrEJ88ch0UQoiII6EWQoiII6EWQoiII6EWQoiII6EWQoiII6EWQoiII6EWQoiII6EWQoiII6EWQoiII6EWQoiII6EWQoiII6EWQlz2JNmnGwVlEkJc9iL91KhCu664c+9oJf735RbbWmFOhRAi4mjpQwghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghIo6EWgghLNr8/4+IqGB9KzrFAAAAAElFTkSuQmCC", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "ref: \n", "- Escribe siempre al menos un párrafo de introducción en cada capítulo o\n", "apartado, explicando de qué vas a tratar en esa sección. Evita que\n", "aparezcan dos encabezados de nivel consecutivos sin ningún texto entre\n", "medias.\n", " Repasa las citas bibliográficas. Comprueba que todas ellas son correctas y siguen\n", "la normativa que exige la titulación.\n", " Asegúrate de que las figuras y las tablas se ven clara y correctamente, e incluyen\n", "número y título, así como su procedencia o fuente.\n", " Comprueba que los índices se generan correctamente.\n", "1.5. Normativa de citas\n", "En esta titulación se cita de acuerdo con la normativa APA.\n", "Recuerda que tienes una guía con explicaciones y ejemplos en el apartado Citas y\n", "bibliografía del aula virtual: https://bibliografiaycitas.unir.net/\n", "© Universidad Internacional de La Rioja (UNIR)\n", "Instrucciones para la redacción y elaboración del TFE\n", "8\n", "Máster Universitario en Inteligencia Artificial\n", "paddle_text: \n", "Escribe siempre al menos un párrafo de introducción en cada capítulo o\n", "apartado,explicando de qué vas a tratar en esa sección. Evita que\n", "aparezcan dos encabezados de nivel consecutivos sin ningún texto entre\n", "medias.\n", "Repasa las citas bibliográficas. Comprueba que todas ellas son correctas y siguen\n", "la normativa que exige la titulación.\n", "Asegúrate de que las figuras y las tablas se ven clara y correctamente, e incluyen\n", "número y título, así como su procedencia o fuente.\n", "Comprueba que los índices se generan correctamente.\n", "1.5. Normativa adecitas\n", "En esta titulacióon se cita de acuerdo con la normativa Apa.\n", "Recuerda que tienes una guía con explicaciones y ejemplos en el apartado Citas y\n", "bibliografía del aula virtual: https://bibliografiaycitas.unir.net/\n", "© Universidad Internacional de La Rioja (UNIR)\n", "Instrucciones para la redacción y elaboración del TfE\n", "Máster Universitario en lnteligencia Artificial ∞\n" ] }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "ref: \n", "2. Estructura del documento\n", "En esta sección se describe con mayor profundidad la estructura y los contenidos\n", "esperados en cada apartado de tu TFE.\n", "Léela con detenimiento y compárala con la programación semanal que encontrarás\n", "en el aula virtual, pues en cada borrador deberás entregar completados diferentes\n", "apartados que se explican a continuación, y que se elaboran de una manera no\n", "necesariamente lineal.\n", "Como ya se ha mencionado, la memoria debe estar estructurada en capítulos. Por\n", "norma general, la estructura de capítulos suele reflejar la línea de discurso del\n", "trabajo, empezando por una introducción donde se plantea el problema, seguida de\n", "un estudio de la literatura donde se estudia y describe el contexto. Posteriormente\n", "se establecen claramente la hipótesis de trabajo y los objetivos concretos de\n", "investigación, así como la descripción de la metodología seguida para alcanzar los\n", "objetivos. Posteriormente se describe la contribución del trabajo, seguida de una\n", "evaluación de la misma. La evaluación da pie a la elaboración de las conclusiones,\n", "que deben relacionar los resultados obtenidos con los objetivos planteados\n", "inicialmente. Finalmente, se describen las líneas de trabajo futuro necesarias para\n", "seguir avanzando hacia la consecución de los objetivos.\n", "A continuación, te dejamos algunos consejos generales sobre cómo organizar los\n", "capítulos, pero ten en cuenta que cada trabajo es único y esta organización es una\n", "guía general adaptable. El director específico de tu TFE podrá aportarte consejos\n", "sobre cómo organizar la memoria adaptándote al contexto de tu trabajo concreto.\n", "Como recomendación general, la estructura de capítulos de tu memoria debería ser\n", "similar a la siguiente propuesta:\n", "© Universidad Internacional de La Rioja (UNIR)\n", " Organización del trabajo en grupo (solo en trabajos grupales)\n", " Capítulo 1 – Introducción\n", "Instrucciones para la redacción y elaboración del TFE\n", "9\n", "Máster Universitario en Inteligencia Artificial\n", "paddle_text: \n", "2.E Estructura del documento\n", "En esta sección se describe con mayor profundidad la estructura y los contenidos\n", "esperados en cada apartado de tu Tfe.\n", "Léela con detenimiento y compárala con la programación semanal que encontraras\n", "en el aula virtual, pues en cada borrador deberás entregar completados diferentes\n", "apartados que se explican a continuación,y que se elaboran de una manera no\n", "necesariamente lineal.\n", "Como ya se ha mencionado, la memoria debe estar estructurada en capítulos. Por\n", "norma general, la estructura de capitulos suele reflejar la linea de discurso del\n", "trabajo, empezando por una introducción donde se plantea el problema, seguida de\n", "un estudio de la literatura donde se estudia y describe el contexto. Posteriormente\n", "se establecen claramente la hipótesis de trabajo y los objetivos concretos de\n", "investigación, así como la descripción de la metodología seguida para alcanzar los\n", "objetivos. Posteriormente se describe la contribución del trabajo, seguida de una\n", "evaluación de la misma. La evaluación da pie a la elaboración de las conclusiones,\n", "que deben relacionar los resultados obtenidos con los objetivos planteados\n", "inicialmente. Finalmente, se describen las líneas de trabajo futuro necesarias para\n", "seguir avanzando hacia la consecución de los objetivos.\n", "A continuación, te dejamos algunos consejos generales sobre cómo organizar los\n", "capítulos, pero ten en cuenta que cada trabajo es único y esta organización es una\n", "guía general adaptable. El director especifico de tu TFE podrá aportarte consejos\n", "sobre cómo organizar la memoria adaptándote al contexto de tu trabajo concreto.\n", "Como recomendación general, la estructura de capítulos de tu memoria debería ser\n", "similar a la siguiente propuesta:\n", "© Universidad Internacional de La Rioja (UNiR)\n", "Organización del trabajo en grupo (solo en trabajos grupales)\n", "Capítulo1–Introducción\n", "Instrucciones para la redacción y elaboración del TFE\n", "Máster Universitario en Inteligencia Artificial 6\n" ] }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "ref: \n", "Capítulo 2 – Contexto y estado del arte\n", " Capítulo 3 – Objetivos y metodología de trabajo\n", " Capítulo 4 – capítulo de desarrollo de la contribución, título del capítulo\n", "dependiendo de la tipología del trabajo\n", " Capítulo 5 – capítulo de desarrollo de la contribución, título del capítulo\n", "dependiendo de la tipología del trabajo\n", " Capítulo 6 – capítulo de desarrollo de la contribución, título del capítulo\n", "dependiendo de la tipología del trabajo\n", " Capítulo 7 – Conclusiones y trabajo futuro\n", "2.1. Resumen\n", "El resumen se redacta en último lugar ya que recoge las contribuciones más\n", "importantes del trabajo. Es necesario tener muy clara y completa del documento para\n", "poder resumirlo correctamente.\n", "Tendrá una extensión de 150 a 300 palabras y deberá ofrecer una visión global de lo\n", "que el lector encontrará en el trabajo, destacando sus aspectos fundamentales.\n", "Deberás indicar claramente cuál es el objetivo principal del trabajo, la metodología\n", "seguida para alcanzarlo, los resultados obtenidos y la principal conclusión alcanzada.\n", "A continuación, indicarás de 3 a 5 palabras clave o keywords como descriptores del\n", "trabajo que lo enmarcan en unas temáticas determinadas. Serán los utilizados para\n", "localizar tu trabajo si llega a ser publicado.\n", "© Universidad Internacional de La Rioja (UNIR)\n", "Instrucciones para la redacción y elaboración del TFE\n", "10\n", "Máster Universitario en Inteligencia Artificial\n", "paddle_text: \n", "Capitulo 2 – Contexto y estado del arte\n", "Capítulo 3 – Objetivos y metodología de trabajo\n", "Capítulo 4 – capítulo de desarrollo de la contribución, título del capítulo\n", "dependiendo de la tipología del trabajo\n", "Capítulo 5 – capítulo de desarrollo de la contribución, título del capítulo\n", "dependiendo de la tipología del trabajo\n", "Capítulo 6 – capítulo de desarrollo de la contribución, título del capítulo\n", "dependiendo de la tipología del trabajo\n", "Capítulo 7 – Conclusiones y trabajo futuro\n", "2.1. Resumen\n", "El resumen se redacta en último lugar ya que recoge las contribuciones más\n", "importantes del trabajo. Es necesario tener muy clara y completa del documento para\n", "poder resumirlo correctamente.\n", "Tendrá una extensión de 150 a 300 palabras y deberá ofrecer una visión global de lo\n", "que el lector encontrará en el trabajo,destacando sus aspectos fundamentales.\n", "Deberás indicar claramente cuál es el objetivo principal del trabajo, la metodología\n", "seguida para alcanzarlo, los resultados obtenidos y la principal conclusión alcanzada.\n", "A continuación, indicarás de 3 a 5 palabras clave o keywords como descriptores del\n", "trabajo que lo enmarcan en unas temáticas determinadas. Serán los utilizados para\n", "localizar tu trabajo si llega a ser publicado.\n", "© Universidad Internacional de La Rioja (UNIR)\n", "Instrucciones para la redacción y elaboración del TFE 10\n", "Máster Universitario en lnteligencia Artificial\n" ] } ], "source": [ "from itertools import islice\n", "\n", "results = []\n", "for img, txt in islice(dataset, 5, 10):\n", " image_array = np.array(img)\n", " out = paddleocr_model.predict(\n", " image_array,\n", " use_doc_orientation_classify=False,\n", " use_doc_unwarping=False,\n", " use_textline_orientation=True\n", " )\n", " show_page(img, 0.15)\n", " print(f\"ref: \\n{txt}\")\n", " paddle_text = assemble_from_paddle_result(out)\n", " print(f\"paddle_text: \\n{paddle_text}\")\n", " results.append({'Model': 'PaddleOCR', 'Prediction': paddle_text, **evaluate_text(txt, paddle_text)})\n", " " ] }, { "cell_type": "markdown", "id": "0db6dc74", "metadata": {}, "source": [ "## 5 Save and Analyze Results" ] }, { "cell_type": "code", "execution_count": 15, "id": "da3155e3", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Benchmark results saved as ai_ocr_benchmark_finetune_results_20251207_155752.csv\n", " WER CER\n", "Model \n", "PaddleOCR 0.104067 0.012581\n" ] }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "df_results = pd.DataFrame(results)\n", "\n", "# Generate a unique filename with timestamp\n", "timestamp = datetime.now().strftime(\"%Y%m%d_%H%M%S\")\n", "filename = f\"ai_ocr_benchmark_finetune_results_{timestamp}.csv\"\n", "filepath = os.path.join(OUTPUT_FOLDER, filename)\n", "\n", "df_results.to_csv(filepath, index=False)\n", "print(f\"Benchmark results saved as {filename}\")\n", "\n", "# Summary by model\n", "summary = df_results.groupby('Model')[['WER', 'CER']].mean()\n", "print(summary)\n", "\n", "# Plot\n", "summary.plot(kind='bar', figsize=(8,5), title='AI OCR Benchmark (WER & CER)')\n", "plt.ylabel('Error Rate')\n", "plt.show()" ] }, { "cell_type": "markdown", "id": "3e0f00c0", "metadata": {}, "source": [ "### How to read this chart:\n", "- CER (Character Error Rate) focus on raw transcription quality\n", "- WER (Word Error Rate) penalizes incorrect tokenization or missing spaces\n", "- CER and WER are error metrics, which means:\n", " - Higher values = worse performance\n", " - Lower values = better accuracy" ] }, { "cell_type": "markdown", "id": "830b0e25", "metadata": {}, "source": [ "# Busqueda de hyperparametros\n", "https://docs.ray.io/en/latest/tune/index.html" ] }, { "cell_type": "code", "execution_count": 16, "id": "3a4bd700", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Python 3.11.9\n", "pip 25.3 from c:\\Users\\Sergio\\Desktop\\MastersThesis\\.venv\\Lib\\site-packages\\pip (python 3.11)\n", "\n" ] } ], "source": [ "!python --version\n", "!pip --version" ] }, { "cell_type": "code", "execution_count": 17, "id": "b0cf4bcf", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Requirement already satisfied: ray[tune] in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (2.52.1)\n", "Requirement already satisfied: click!=8.3.*,>=7.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ray[tune]) (8.2.1)\n", "Requirement already satisfied: filelock in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ray[tune]) (3.20.0)\n", "Requirement already satisfied: jsonschema in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ray[tune]) (4.25.1)\n", "Requirement already satisfied: msgpack<2.0.0,>=1.0.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ray[tune]) (1.1.2)\n", "Requirement already satisfied: packaging in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ray[tune]) (25.0)\n", "Requirement already satisfied: protobuf>=3.20.3 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ray[tune]) (6.33.2)\n", "Requirement already satisfied: pyyaml in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ray[tune]) (6.0.2)\n", "Requirement already satisfied: requests in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ray[tune]) (2.32.5)\n", "Requirement already satisfied: pandas in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ray[tune]) (2.3.3)\n", "Requirement already satisfied: pydantic!=2.0.*,!=2.1.*,!=2.2.*,!=2.3.*,!=2.4.*,<3 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ray[tune]) (2.12.5)\n", "Requirement already satisfied: tensorboardX>=1.9 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ray[tune]) (2.6.4)\n", "Requirement already satisfied: pyarrow>=9.0.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ray[tune]) (22.0.0)\n", "Requirement already satisfied: fsspec in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from ray[tune]) (2025.12.0)\n", "Requirement already satisfied: annotated-types>=0.6.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from pydantic!=2.0.*,!=2.1.*,!=2.2.*,!=2.3.*,!=2.4.*,<3->ray[tune]) (0.7.0)\n", "Requirement already satisfied: pydantic-core==2.41.5 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from pydantic!=2.0.*,!=2.1.*,!=2.2.*,!=2.3.*,!=2.4.*,<3->ray[tune]) (2.41.5)\n", "Requirement already satisfied: typing-extensions>=4.14.1 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from pydantic!=2.0.*,!=2.1.*,!=2.2.*,!=2.3.*,!=2.4.*,<3->ray[tune]) (4.15.0)\n", "Requirement already satisfied: typing-inspection>=0.4.2 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from pydantic!=2.0.*,!=2.1.*,!=2.2.*,!=2.3.*,!=2.4.*,<3->ray[tune]) (0.4.2)\n", "Requirement already satisfied: colorama in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from click!=8.3.*,>=7.0->ray[tune]) (0.4.6)\n", "Requirement already satisfied: numpy in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from tensorboardX>=1.9->ray[tune]) (2.3.5)\n", "Requirement already satisfied: attrs>=22.2.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jsonschema->ray[tune]) (25.4.0)\n", "Requirement already satisfied: jsonschema-specifications>=2023.03.6 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jsonschema->ray[tune]) (2025.9.1)\n", "Requirement already satisfied: referencing>=0.28.4 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jsonschema->ray[tune]) (0.37.0)\n", "Requirement already satisfied: rpds-py>=0.7.1 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from jsonschema->ray[tune]) (0.30.0)\n", "Requirement already satisfied: python-dateutil>=2.8.2 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from pandas->ray[tune]) (2.9.0.post0)\n", "Requirement already satisfied: pytz>=2020.1 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from pandas->ray[tune]) (2025.2)\n", "Requirement already satisfied: tzdata>=2022.7 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from pandas->ray[tune]) (2025.2)\n", "Requirement already satisfied: six>=1.5 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from python-dateutil>=2.8.2->pandas->ray[tune]) (1.17.0)\n", "Requirement already satisfied: charset_normalizer<4,>=2 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from requests->ray[tune]) (3.4.4)\n", "Requirement already satisfied: idna<4,>=2.5 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from requests->ray[tune]) (3.11)\n", "Requirement already satisfied: urllib3<3,>=1.21.1 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from requests->ray[tune]) (2.6.0)\n", "Requirement already satisfied: certifi>=2017.4.17 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from requests->ray[tune]) (2025.11.12)\n", "Note: you may need to restart the kernel to use updated packages.\n", "Requirement already satisfied: optuna in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (4.6.0)\n", "Requirement already satisfied: alembic>=1.5.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from optuna) (1.17.2)\n", "Requirement already satisfied: colorlog in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from optuna) (6.10.1)\n", "Requirement already satisfied: numpy in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from optuna) (2.3.5)\n", "Requirement already satisfied: packaging>=20.0 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from optuna) (25.0)\n", "Requirement already satisfied: sqlalchemy>=1.4.2 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from optuna) (2.0.44)\n", "Requirement already satisfied: tqdm in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from optuna) (4.67.1)\n", "Requirement already satisfied: PyYAML in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from optuna) (6.0.2)\n", "Requirement already satisfied: Mako in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from alembic>=1.5.0->optuna) (1.3.10)\n", "Requirement already satisfied: typing-extensions>=4.12 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from alembic>=1.5.0->optuna) (4.15.0)\n", "Requirement already satisfied: greenlet>=1 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from sqlalchemy>=1.4.2->optuna) (3.3.0)\n", "Requirement already satisfied: colorama in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from colorlog->optuna) (0.4.6)\n", "Requirement already satisfied: MarkupSafe>=0.9.2 in c:\\users\\sergio\\desktop\\mastersthesis\\.venv\\lib\\site-packages (from Mako->alembic>=1.5.0->optuna) (3.0.3)\n", "Note: you may need to restart the kernel to use updated packages.\n" ] } ], "source": [ "# Instalación de Ray y Ray Tune\n", "%pip install -U \"ray[tune]\" \n", "%pip install optuna" ] }, { "cell_type": "code", "execution_count": 6, "id": "ae5a10c4", "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "2025-12-07 19:58:07,710\tINFO worker.py:2023 -- Started a local Ray instance.\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Ray Tune listo (versión: 2.52.1 )\n" ] }, { "name": "stderr", "output_type": "stream", "text": [ "c:\\Users\\Sergio\\Desktop\\MastersThesis\\.venv\\Lib\\site-packages\\ray\\_private\\worker.py:2062: FutureWarning: Tip: In future versions of Ray, Ray will no longer override accelerator visible devices env var if num_gpus=0 or num_gpus=None (default). To enable this behavior and turn off this error message, set RAY_ACCEL_ENV_VAR_OVERRIDE_ON_ZERO=0\n", " warnings.warn(\n" ] } ], "source": [ "# ===============================================================\n", "# 🔍 RAY TUNE: OPTIMIZACIÓN AUTOMÁTICA DE HIPERPARÁMETROS OCR\n", "# ===============================================================\n", "import ray\n", "from ray import tune, air\n", "import pandas as pd\n", "\n", "\n", "ray.init(ignore_reinit_error=True)\n", "print(\"Ray Tune listo (versión:\", ray.__version__, \")\")\n" ] }, { "cell_type": "code", "execution_count": 19, "id": "96c320e8", "metadata": {}, "outputs": [], "source": [ "# --- Configuración base del experimento ---\n", "search_space = {\n", " #Whether to use document image orientation classification.\n", " \"use_doc_orientation_classify\": tune.choice([True, False]), \n", " # Whether to use text image unwarping.\n", " \"use_doc_unwarping\": tune.choice([True, False]),\n", " # Whether to use text line orientation classification.\n", " \"textline_orientation\": tune.choice([True, False]),\n", " # Detection pixel threshold for the text detection model. Pixels with scores greater than this threshold in the output probability map are considered text pixels.\n", " \"text_det_thresh\" : tune.uniform(0.0, 0.7),\n", " # Detection box threshold for the text detection model. A detection result is considered a text region if the average score of all pixels within the border of the result is greater than this threshold.\n", " \"text_det_box_thresh\": tune.uniform(0.0, 0.7),\n", " # Text detection expansion coefficient, which expands the text region using this method. The larger the value, the larger the expansion area.\n", " \"text_det_unclip_ratio\": tune.choice([0.0]),\n", " # Text recognition threshold. Text results with scores greater than this threshold are retained.\n", " \"text_rec_score_thresh\": tune.uniform(0.0, 0.7),\n", "}\n", "KEYMAP = {\n", " \"textline_orientation\": \"textline-orientation\",\n", " \"use_doc_unwarping\": \"use-doc-unwarping\",\n", " \"use_doc_orientation_classify\": \"use-doc-orientation-classify\",\n", " \"text_det_box_thresh\": \"text-det-box-thresh\",\n", " \"text_det_unclip_ratio\": \"text-det-unclip-ratio\",\n", " \"text_rec_score_thresh\": \"text-rec-score-thresh\",\n", " \"text_det_thresh\": \"text-det-thresh\"\n", "}" ] }, { "cell_type": "code", "execution_count": 20, "id": "accb4e9d", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Notebook Python: c:\\Users\\Sergio\\Desktop\\MastersThesis\\.venv\\Scripts\\python.exe\n" ] }, { "name": "stderr", "output_type": "stream", "text": [ "\u001b[36m(pid=gcs_server)\u001b[0m [2025-12-07 15:58:31,070 E 25184 15184] (gcs_server.exe) gcs_server.cc:303: Failed to establish connection to the event+metrics exporter agent. Events and metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\n", "\u001b[33m(raylet)\u001b[0m [2025-12-07 15:58:32,657 E 10072 20448] (raylet.exe) main.cc:979: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\n", "\u001b[36m(pid=18776)\u001b[0m [2025-12-07 15:58:36,373 E 18776 26484] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "{'CER': 0.012581110635031723, 'WER': 0.10406694286511942, 'TIME': 331.0908589363098, 'PAGES': 5, 'TIME_PER_PAGE': 66.11821403503419}\n", "return code: 0\n", "args: ['c:\\\\Users\\\\Sergio\\\\Desktop\\\\MastersThesis\\\\.venv\\\\Scripts\\\\python.exe', 'c:\\\\Users\\\\Sergio\\\\Desktop\\\\MastersThesis\\\\src\\\\paddle_ocr_tuning.py', '--pdf-folder', 'c:\\\\Users\\\\Sergio\\\\Desktop\\\\MastersThesis\\\\src\\\\dataset', '--textline-orientation', 'True', '--use-doc-unwarping', 'False', '--use-doc-orientation-classify', 'False', '--text-det-box-thresh', '0.0', '--text-det-unclip-ratio', '1.5', '--text-det-thresh', '0.0', '--text-rec-score-thresh', '0.0']\n" ] } ], "source": [ "import sys, subprocess\n", "print(\"Notebook Python:\", sys.executable)\n", "# test paddle ocr run with params\n", "args = [sys.executable, \n", " SCRIPT_ABS, \n", " \"--pdf-folder\", PDF_FOLDER_ABS, \n", " \"--textline-orientation\",\"True\",\n", " \"--use-doc-unwarping\",\"False\",\n", " \"--use-doc-orientation-classify\",\"False\",\n", " \"--text-det-box-thresh\",\"0.0\",\n", " \"--text-det-unclip-ratio\",\"1.5\",\n", " \"--text-det-thresh\", \"0.0\",\n", " \"--text-rec-score-thresh\",\"0.0\"]\n", "test_proc = subprocess.run(args, capture_output=True, text=True, cwd=SCRIPT_DIR)\n", "if test_proc.returncode != 0:\n", " print(test_proc.stderr)\n", "last = test_proc.stdout.strip().splitlines()[-1]\n", "\n", "metrics = json.loads(last)\n", "print(metrics)\n", "\n", "print(f\"return code: {test_proc.returncode}\")\n", "print(f\"args: {args}\")" ] }, { "cell_type": "code", "execution_count": 21, "id": "8df28468", "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "c:\\Users\\Sergio\\Desktop\\MastersThesis\\.venv\\Lib\\site-packages\\ray\\tune\\impl\\tuner_internal.py:144: RayDeprecationWarning: The `RunConfig` class should be imported from `ray.tune` when passing it to the Tuner. Please update your imports. See this issue for more context and migration options: https://github.com/ray-project/ray/issues/49454. Disable these warnings by setting the environment variable: RAY_TRAIN_ENABLE_V2_MIGRATION_WARNINGS=0\n", " _log_deprecation_warning(\n", "2025-12-07 16:03:56,654\tINFO tune.py:616 -- [output] This uses the legacy output and progress reporter, as Jupyter notebooks are not supported by the new engine, yet. For more information, please see https://github.com/ray-project/ray/issues/36949\n", "[I 2025-12-07 16:03:56,662] A new study created in memory with name: optuna\n" ] }, { "data": { "text/html": [ "
\n", "
\n", "
\n", "

Tune Status

\n", " \n", "\n", "\n", "\n", "\n", "\n", "
Current time:2025-12-07 19:23:17
Running for: 03:19:21.23
Memory: 4.4/15.9 GiB
\n", "
\n", "
\n", "
\n", "

System Info

\n", " Using FIFO scheduling algorithm.
Logical resource usage: 1.0/16 CPUs, 0/1 GPUs (0.0/1.0 accelerator_type:G)\n", "
\n", " \n", "
\n", "
\n", "
\n", "

Trial Status

\n", " \n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "
Trial name status loc text_det_box_thresh text_det_thresh text_det_unclip_rati\n", "o text_rec_score_thres\n", "htextline_orientation use_doc_orientation_\n", "classify use_doc_unwarping iter total time (s) CER WER TIME
trainable_paddle_ocr_d5238c33TERMINATED127.0.0.1:19452 0.623029 0.088782100.229944 True True False 1 374.2780.01351590.105003 353.851
trainable_paddle_ocr_ea8a2f7aTERMINATED127.0.0.1:7472 0.671201 0.393201 00.168802 False FalseFalse 1 374.3 0.039052 0.132086 354.615
trainable_paddle_ocr_ebb12e5bTERMINATED127.0.0.1:21480 0.235725 0.432878 00.184435 True True True 1 379.5440.06606240.166192 359.097
trainable_paddle_ocr_b3775034TERMINATED127.0.0.1:23084 0.337744 0.064128800.576405 False True True 1 356.5260.418109 0.50371 336.661
trainable_paddle_ocr_bf10d370TERMINATED127.0.0.1:26140 0.690232 0.671955 00.39649 True True True 1 370.9030.197252 0.295353 350.147
trainable_paddle_ocr_111e5a9eTERMINATED127.0.0.1:20664 0.483266 0.044816 00.546416 False True False 1 341.0710.38641 0.455836 320.966
trainable_paddle_ocr_415d7ba1TERMINATED127.0.0.1:23848 0.523385 0.016997100.208331 True True True 1 347.2990.516069 0.59453 326.657
trainable_paddle_ocr_a58d8109TERMINATED127.0.0.1:25248 0.670589 0.040243200.188585 True FalseTrue 1 346.09 0.502513 0.567716 326.916
trainable_paddle_ocr_33bdf2a9TERMINATED127.0.0.1:24024 0.490009 0.434737 00.151906 False FalseTrue 1 388.1510.07092030.17391 368.571
trainable_paddle_ocr_d9df79f3TERMINATED127.0.0.1:5368 0.626194 0.178064 00.385477 False True True 1 384.6770.116825 0.22213 364.623
trainable_paddle_ocr_80ea65f2TERMINATED127.0.0.1:14064 0.251382 0.601112 00.313124 False True True 1 387.6790.06459480.164937 366.607
trainable_paddle_ocr_2e978bfaTERMINATED127.0.0.1:11060 0.0777319 0.234859 00.0236948 True FalseFalse 1 380.2810.01340060.107419 359.597
trainable_paddle_ocr_8518cc40TERMINATED127.0.0.1:21016 0.000241868 0.222556 00.00289108True FalseFalse 1 368.5460.01340060.107419 347.929
trainable_paddle_ocr_2c691aaaTERMINATED127.0.0.1:21540 0.0303334 0.224727 00.0509969 True FalseFalse 1 366.3460.01340060.107419 347.145
trainable_paddle_ocr_31e60691TERMINATED127.0.0.1:17532 0.00196041 0.259141 00.00350944True FalseFalse 1 368.0380.01304040.104854 347.22
trainable_paddle_ocr_d4d288c6TERMINATED127.0.0.1:22216 0.00339892 0.273408 00.0154205 True FalseFalse 1 368.9040.01258290.10328 349.232
trainable_paddle_ocr_7645b77cTERMINATED127.0.0.1:2272 0.113841 0.279242 00.0753151 True FalseFalse 1 367.4560.01258290.10328 346.698
trainable_paddle_ocr_3256ae36TERMINATED127.0.0.1:6604 0.129213 0.30993 00.11202 True FalseFalse 1 366.0020.01240760.102016 346.52
trainable_paddle_ocr_b0dda58bTERMINATED127.0.0.1:9732 0.117838 0.314952 00.682573 True FalseFalse 1 364.8280.01240760.102016 344.029
trainable_paddle_ocr_e9d40333TERMINATED127.0.0.1:23416 0.156939 0.530252 00.100194 True FalseFalse 1 365.6260.01242980.102051 346.118
trainable_paddle_ocr_aa89fe7aTERMINATED127.0.0.1:16200 0.162083 0.50397 00.676539 True FalseFalse 1 366.7530.01199070.100476 346.54
trainable_paddle_ocr_92c48d07TERMINATED127.0.0.1:15432 0.186443 0.333219 00.67753 True FalseFalse 1 365.0940.01196850.100441 345.979
trainable_paddle_ocr_187790d7TERMINATED127.0.0.1:24676 0.235252 0.337251 00.698732 True FalseFalse 1 364.4740.01196850.100441 344.173
trainable_paddle_ocr_442a2439TERMINATED127.0.0.1:7892 0.212276 0.509804 00.699247 True FalseFalse 1 364.7550.01176010.0996499345.943
trainable_paddle_ocr_70862adcTERMINATED127.0.0.1:15412 0.216306 0.396397 00.685918 True FalseFalse 1 365.9750.01196850.100441 345.403
trainable_paddle_ocr_e6821f34TERMINATED127.0.0.1:26088 0.240775 0.366898 00.573762 True FalseFalse 1 365.2550.01240760.102016 345.881
trainable_paddle_ocr_8b680875TERMINATED127.0.0.1:1720 0.319343 0.53125 00.591253 True FalseFalse 1 367.2030.01219920.101225 347.056
trainable_paddle_ocr_fc54867bTERMINATED127.0.0.1:4888 0.304286 0.503408 00.502491 True FalseFalse 1 368.7360.01242980.102051 349.607
trainable_paddle_ocr_c32d0d5eTERMINATED127.0.0.1:25808 0.398489 0.153007 00.516768 True FalseFalse 1 364.4230.01338550.109273 343.855
trainable_paddle_ocr_4762fbbbTERMINATED127.0.0.1:20760 0.40101 0.133426 00.618812 True FalseFalse 1 363.3260.01353720.108525 344.601
trainable_paddle_ocr_522ac97cTERMINATED127.0.0.1:2372 0.402755 0.448976 00.642637 True FalseFalse 1 364.72 0.01176380.099689 344.038
trainable_paddle_ocr_5784f433TERMINATED127.0.0.1:22900 0.192769 0.46205 00.632828 True FalseFalse 1 362.93 0.01165030.0989016343.513
trainable_paddle_ocr_83af0528TERMINATED127.0.0.1:9832 0.184587 0.466314 00.629921 True FalseFalse 1 364.5850.01165030.0989016343.81
trainable_paddle_ocr_12cbaa22TERMINATED127.0.0.1:5968 0.405622 0.472779 00.631499 True FalseFalse 1 364.2470.01165030.0989016344.114
trainable_paddle_ocr_a3a87765TERMINATED127.0.0.1:24372 0.28557 0.4501 00.635152 True FalseFalse 1 369.2740.01176380.099689 348.58
trainable_paddle_ocr_cf2bad0cTERMINATED127.0.0.1:3272 0.283661 0.589012 00.460291 False FalseFalse 1 366.1880.044199 0.132047 347.034
trainable_paddle_ocr_9a9b91e7TERMINATED127.0.0.1:2272 0.364609 0.608959 00.465225 False FalseFalse 1 364.0170.044199 0.132047 343.539
trainable_paddle_ocr_e326d901TERMINATED127.0.0.1:24932 0.373537 0.593229 00.463688 True FalseFalse 1 365.4280.01219920.101225 345.762
trainable_paddle_ocr_ccb3f19aTERMINATED127.0.0.1:1104 0.453777 0.686641 00.305928 True True False 1 365.1470.01199030.0991043344.408
trainable_paddle_ocr_8c12c55fTERMINATED127.0.0.1:19700 0.444416 0.67104 00.264132 True True False 1 363.2970.01218620.101228 343.939
trainable_paddle_ocr_5a62d5b6TERMINATED127.0.0.1:26528 0.201047 0.404141 00.599257 True True True 1 380.3330.06627090.168515 359.467
trainable_paddle_ocr_bb4495b7TERMINATED127.0.0.1:21772 0.576439 0.390737 00.541396 False FalseTrue 1 375.9770.07070080.17391 356.322
trainable_paddle_ocr_9d90711dTERMINATED127.0.0.1:17592 0.541158 0.468954 00.635015 True FalseFalse 1 365.77 0.01153510.0989016344.718
trainable_paddle_ocr_daaec3f8TERMINATED127.0.0.1:21292 0.521341 0.474351 00.644567 True FalseFalse 1 363.0190.01153510.0989016343.697
trainable_paddle_ocr_51fb5915TERMINATED127.0.0.1:21772 0.58105 0.485412 00.64636 True FalseFalse 1 364.02 0.01153510.0989016343.604
trainable_paddle_ocr_18966a33TERMINATED127.0.0.1:16900 0.51329 0.550159 00.648982 True FalseFalse 1 363.3370.01164490.0996499344.261
trainable_paddle_ocr_b67080f9TERMINATED127.0.0.1:20948 0.576074 0.553412 00.560972 True FalseFalse 1 366.0190.01231450.102051 345.495
trainable_paddle_ocr_2533f368TERMINATED127.0.0.1:11208 0.524608 0.557227 00.558307 True FalseTrue 1 371.2050.07209120.179189 351.967
trainable_paddle_ocr_451d018dTERMINATED127.0.0.1:3616 0.549464 0.634019 00.652105 False FalseTrue 1 378.8270.06479950.164937 357.17
trainable_paddle_ocr_2256e752TERMINATED127.0.0.1:25468 0.622863 0.647804 00.654609 False True False 1 369.88 0.04429210.132838 349.417
trainable_paddle_ocr_0a892729TERMINATED127.0.0.1:26212 0.542929 0.421733 00.601587 True FalseFalse 1 367.2370.01229230.102016 346.072
trainable_paddle_ocr_495075f5TERMINATED127.0.0.1:23604 0.631875 0.418675 00.595618 True FalseFalse 1 365.5360.01229230.102016 346.425
trainable_paddle_ocr_54c45552TERMINATED127.0.0.1:25352 0.619687 0.463823 00.612612 True FalseFalse 1 367.9470.01197420.100476 346.941
trainable_paddle_ocr_6b2e9b93TERMINATED127.0.0.1:25400 0.48925 0.475185 00.515482 True FalseFalse 1 365.9890.01197420.100476 346.414
trainable_paddle_ocr_e9a6b81fTERMINATED127.0.0.1:4036 0.492552 0.48793 00.648349 True FalseFalse 1 367.3320.01153510.0989016346.259
trainable_paddle_ocr_076c5450TERMINATED127.0.0.1:4832 0.588133 0.488422 00.656919 True FalseFalse 1 365.1880.01153510.0989016345.843
trainable_paddle_ocr_4a42a3eaTERMINATED127.0.0.1:14912 0.594041 0.559036 00.657323 True FalseFalse 1 370.9970.01187540.100476 350.244
trainable_paddle_ocr_041795f1TERMINATED127.0.0.1:22372 0.661744 0.565009 00.66295 True FalseFalse 1 370.9460.01208010.100476 351.5
trainable_paddle_ocr_8abb3f37TERMINATED127.0.0.1:22012 0.463682 0.489821 00.394583 True FalseFalse 1 364.6750.01231450.102051 343.539
trainable_paddle_ocr_f2cb682eTERMINATED127.0.0.1:5752 0.452248 0.491795 00.425971 True True False 1 364.9080.01231450.102051 345.592
trainable_paddle_ocr_463fe5e7TERMINATED127.0.0.1:16524 0.520238 0.537344 00.534057 True True False 1 370.5640.01231450.102051 349.509
trainable_paddle_ocr_88bbe87dTERMINATED127.0.0.1:15084 0.511078 0.527459 00.536896 True FalseFalse 1 369.55 0.01208390.101225 350.144
trainable_paddle_ocr_33ea1cc6TERMINATED127.0.0.1:17380 0.515807 0.522992 00.667966 True FalseFalse 1 376.7460.01187540.100476 355.524
trainable_paddle_ocr_1243723eTERMINATED127.0.0.1:11232 0.557315 0.372677 00.676613 True FalseFalse 1 375.4440.01185320.100441 355.679
\n", "
\n", "
\n", "\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stderr", "output_type": "stream", "text": [ "2025-12-07 16:03:56,713\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_d5238c33_1_text_det_box_thresh=0.6230,text_det_thresh=0.0888,text_det_unclip_ratio=0.0000,text_rec_score_thre_2025-12-07_16-03-56\n", "2025-12-07 16:03:56,718\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_d5238c33_1_text_det_box_thresh=0.6230,text_det_thresh=0.0888,text_det_unclip_ratio=0.0000,text_rec_score_thre_2025-12-07_16-03-56\n", "2025-12-07 16:04:01,625\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_d5238c33_1_text_det_box_thresh=0.6230,text_det_thresh=0.0888,text_det_unclip_ratio=0.0000,text_rec_score_thre_2025-12-07_16-03-56\n", "2025-12-07 16:04:01,626\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_d5238c33_1_text_det_box_thresh=0.6230,text_det_thresh=0.0888,text_det_unclip_ratio=0.0000,text_rec_score_thre_2025-12-07_16-03-56\n", "2025-12-07 16:04:01,639\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_ea8a2f7a_2_text_det_box_thresh=0.6712,text_det_thresh=0.3932,text_det_unclip_ratio=0.0000,text_rec_score_thre_2025-12-07_16-04-01\n", "2025-12-07 16:04:01,642\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_ea8a2f7a_2_text_det_box_thresh=0.6712,text_det_thresh=0.3932,text_det_unclip_ratio=0.0000,text_rec_score_thre_2025-12-07_16-04-01\n", "2025-12-07 16:04:06,097\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_ea8a2f7a_2_text_det_box_thresh=0.6712,text_det_thresh=0.3932,text_det_unclip_ratio=0.0000,text_rec_score_thre_2025-12-07_16-04-01\n", "2025-12-07 16:04:06,097\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_ea8a2f7a_2_text_det_box_thresh=0.6712,text_det_thresh=0.3932,text_det_unclip_ratio=0.0000,text_rec_score_thre_2025-12-07_16-04-01\n", "\u001b[36m(trainable_paddle_ocr pid=19452)\u001b[0m [2025-12-07 16:04:31,654 E 19452 19604] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\n", "\u001b[36m(trainable_paddle_ocr pid=7472)\u001b[0m [2025-12-07 16:04:37,442 E 7472 7092] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\n" ] }, { "data": { "text/html": [ "
\n", "

Trial Progress

\n", " \n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "
Trial name CER PAGES TIME TIME_PER_PAGE WER
trainable_paddle_ocr_041795f10.0120801 5351.5 70.19010.100476
trainable_paddle_ocr_076c54500.0115351 5345.843 69.06780.0989016
trainable_paddle_ocr_0a8927290.0122923 5346.072 69.12430.102016
trainable_paddle_ocr_111e5a9e0.38641 5320.966 64.09520.455836
trainable_paddle_ocr_1243723e0.0118532 5355.679 71.02430.100441
trainable_paddle_ocr_12cbaa220.0116503 5344.114 68.724 0.0989016
trainable_paddle_ocr_187790d70.0119685 5344.173 68.74230.100441
trainable_paddle_ocr_18966a330.0116449 5344.261 68.75940.0996499
trainable_paddle_ocr_2256e7520.0442921 5349.417 69.77590.132838
trainable_paddle_ocr_2533f3680.0720912 5351.967 70.29540.179189
trainable_paddle_ocr_2c691aaa0.0134006 5347.145 69.32420.107419
trainable_paddle_ocr_2e978bfa0.0134006 5359.597 71.80430.107419
trainable_paddle_ocr_31e606910.0130404 5347.22 69.34550.104854
trainable_paddle_ocr_3256ae360.0124076 5346.52 69.19980.102016
trainable_paddle_ocr_33bdf2a90.0709203 5368.571 73.625 0.17391
trainable_paddle_ocr_33ea1cc60.0118754 5355.524 71.00810.100476
trainable_paddle_ocr_415d7ba10.516069 5326.657 65.23510.59453
trainable_paddle_ocr_442a24390.0117601 5345.943 69.08390.0996499
trainable_paddle_ocr_451d018d0.0647995 5357.17 71.33720.164937
trainable_paddle_ocr_463fe5e70.0123145 5349.509 69.80770.102051
trainable_paddle_ocr_4762fbbb0.0135372 5344.601 68.81450.108525
trainable_paddle_ocr_495075f50.0122923 5346.425 69.19190.102016
trainable_paddle_ocr_4a42a3ea0.0118754 5350.244 69.94840.100476
trainable_paddle_ocr_51fb59150.0115351 5343.604 68.62930.0989016
trainable_paddle_ocr_522ac97c0.0117638 5344.038 68.71830.099689
trainable_paddle_ocr_54c455520.0119742 5346.941 69.29810.100476
trainable_paddle_ocr_5784f4330.0116503 5343.513 68.60030.0989016
trainable_paddle_ocr_5a62d5b60.0662709 5359.467 71.79710.168515
trainable_paddle_ocr_6b2e9b930.0119742 5346.414 69.18590.100476
trainable_paddle_ocr_70862adc0.0119685 5345.403 68.98560.100441
trainable_paddle_ocr_7645b77c0.0125829 5346.698 69.24070.10328
trainable_paddle_ocr_80ea65f20.0645948 5366.607 73.222 0.164937
trainable_paddle_ocr_83af05280.0116503 5343.81 68.66910.0989016
trainable_paddle_ocr_8518cc400.0134006 5347.929 69.49 0.107419
trainable_paddle_ocr_88bbe87d0.0120839 5350.144 69.92810.101225
trainable_paddle_ocr_8abb3f370.0123145 5343.539 68.61340.102051
trainable_paddle_ocr_8b6808750.0121992 5347.056 69.31870.101225
trainable_paddle_ocr_8c12c55f0.0121862 5343.939 68.69270.101228
trainable_paddle_ocr_92c48d070.0119685 5345.979 69.09320.100441
trainable_paddle_ocr_9a9b91e70.044199 5343.539 68.61560.132047
trainable_paddle_ocr_9d90711d0.0115351 5344.718 68.85830.0989016
trainable_paddle_ocr_a3a877650.0117638 5348.58 69.61860.099689
trainable_paddle_ocr_a58d81090.502513 5326.916 65.28340.567716
trainable_paddle_ocr_aa89fe7a0.0119907 5346.54 69.21830.100476
trainable_paddle_ocr_b0dda58b0.0124076 5344.029 68.71350.102016
trainable_paddle_ocr_b37750340.418109 5336.661 67.22690.50371
trainable_paddle_ocr_b67080f90.0123145 5345.495 69.01210.102051
trainable_paddle_ocr_bb4495b70.0707008 5356.322 71.16440.17391
trainable_paddle_ocr_bf10d3700.197252 5350.147 69.93640.295353
trainable_paddle_ocr_c32d0d5e0.0133855 5343.855 68.67560.109273
trainable_paddle_ocr_ccb3f19a0.0119903 5344.408 68.78970.0991043
trainable_paddle_ocr_cf2bad0c0.044199 5347.034 69.311 0.132047
trainable_paddle_ocr_d4d288c60.0125829 5349.232 69.74630.10328
trainable_paddle_ocr_d5238c330.0135159 5353.851 70.66230.105003
trainable_paddle_ocr_d9df79f30.116825 5364.623 72.82480.22213
trainable_paddle_ocr_daaec3f80.0115351 5343.697 68.64240.0989016
trainable_paddle_ocr_e326d9010.0121992 5345.762 69.05780.101225
trainable_paddle_ocr_e6821f340.0124076 5345.881 69.07740.102016
trainable_paddle_ocr_e9a6b81f0.0115351 5346.259 69.15520.0989016
trainable_paddle_ocr_e9d403330.0124298 5346.118 69.12530.102051
trainable_paddle_ocr_ea8a2f7a0.039052 5354.615 70.82210.132086
trainable_paddle_ocr_ebb12e5b0.0660624 5359.097 71.72570.166192
trainable_paddle_ocr_f2cb682e0.0123145 5345.592 69.02380.102051
trainable_paddle_ocr_fc54867b0.0124298 5349.607 69.82530.102051
\n", "
\n", "\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stderr", "output_type": "stream", "text": [ "2025-12-07 16:10:15,969\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_d5238c33_1_text_det_box_thresh=0.6230,text_det_thresh=0.0888,text_det_unclip_ratio=0.0000,text_rec_score_thre_2025-12-07_16-03-56\n", "2025-12-07 16:10:16,056\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_ebb12e5b_3_text_det_box_thresh=0.2357,text_det_thresh=0.4329,text_det_unclip_ratio=0.0000,text_rec_score_thre_2025-12-07_16-10-16\n", "2025-12-07 16:10:16,063\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_ebb12e5b_3_text_det_box_thresh=0.2357,text_det_thresh=0.4329,text_det_unclip_ratio=0.0000,text_rec_score_thre_2025-12-07_16-10-16\n", "2025-12-07 16:10:20,414\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_ea8a2f7a_2_text_det_box_thresh=0.6712,text_det_thresh=0.3932,text_det_unclip_ratio=0.0000,text_rec_score_thre_2025-12-07_16-04-01\n", "2025-12-07 16:10:22,097\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_ebb12e5b_3_text_det_box_thresh=0.2357,text_det_thresh=0.4329,text_det_unclip_ratio=0.0000,text_rec_score_thre_2025-12-07_16-10-16\n", "2025-12-07 16:10:22,097\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_ebb12e5b_3_text_det_box_thresh=0.2357,text_det_thresh=0.4329,text_det_unclip_ratio=0.0000,text_rec_score_thre_2025-12-07_16-10-16\n", "2025-12-07 16:10:22,097\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_b3775034_4_text_det_box_thresh=0.3377,text_det_thresh=0.0641,text_det_unclip_ratio=0.0000,text_rec_score_thre_2025-12-07_16-10-22\n", "2025-12-07 16:10:22,097\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_b3775034_4_text_det_box_thresh=0.3377,text_det_thresh=0.0641,text_det_unclip_ratio=0.0000,text_rec_score_thre_2025-12-07_16-10-22\n", "2025-12-07 16:10:26,662\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_b3775034_4_text_det_box_thresh=0.3377,text_det_thresh=0.0641,text_det_unclip_ratio=0.0000,text_rec_score_thre_2025-12-07_16-10-22\n", "2025-12-07 16:10:26,664\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_b3775034_4_text_det_box_thresh=0.3377,text_det_thresh=0.0641,text_det_unclip_ratio=0.0000,text_rec_score_thre_2025-12-07_16-10-22\n", "\u001b[36m(trainable_paddle_ocr pid=21480)\u001b[0m [2025-12-07 16:10:51,593 E 21480 13444] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\n", "\u001b[36m(trainable_paddle_ocr pid=23084)\u001b[0m [2025-12-07 16:10:56,943 E 23084 15580] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\n", "2025-12-07 16:16:23,218\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_b3775034_4_text_det_box_thresh=0.3377,text_det_thresh=0.0641,text_det_unclip_ratio=0.0000,text_rec_score_thre_2025-12-07_16-10-22\n", "2025-12-07 16:16:23,261\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_bf10d370_5_text_det_box_thresh=0.6902,text_det_thresh=0.6720,text_det_unclip_ratio=0.0000,text_rec_score_thre_2025-12-07_16-16-23\n", "2025-12-07 16:16:23,263\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_bf10d370_5_text_det_box_thresh=0.6902,text_det_thresh=0.6720,text_det_unclip_ratio=0.0000,text_rec_score_thre_2025-12-07_16-16-23\n", "2025-12-07 16:16:28,918\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_bf10d370_5_text_det_box_thresh=0.6902,text_det_thresh=0.6720,text_det_unclip_ratio=0.0000,text_rec_score_thre_2025-12-07_16-16-23\n", "2025-12-07 16:16:28,918\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_bf10d370_5_text_det_box_thresh=0.6902,text_det_thresh=0.6720,text_det_unclip_ratio=0.0000,text_rec_score_thre_2025-12-07_16-16-23\n", "2025-12-07 16:16:41,652\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_ebb12e5b_3_text_det_box_thresh=0.2357,text_det_thresh=0.4329,text_det_unclip_ratio=0.0000,text_rec_score_thre_2025-12-07_16-10-16\n", "2025-12-07 16:16:41,663\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_111e5a9e_6_text_det_box_thresh=0.4833,text_det_thresh=0.0448,text_det_unclip_ratio=0.0000,text_rec_score_thre_2025-12-07_16-16-41\n", "2025-12-07 16:16:41,665\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_111e5a9e_6_text_det_box_thresh=0.4833,text_det_thresh=0.0448,text_det_unclip_ratio=0.0000,text_rec_score_thre_2025-12-07_16-16-41\n", "2025-12-07 16:16:46,207\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_111e5a9e_6_text_det_box_thresh=0.4833,text_det_thresh=0.0448,text_det_unclip_ratio=0.0000,text_rec_score_thre_2025-12-07_16-16-41\n", "2025-12-07 16:16:46,207\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_111e5a9e_6_text_det_box_thresh=0.4833,text_det_thresh=0.0448,text_det_unclip_ratio=0.0000,text_rec_score_thre_2025-12-07_16-16-41\n", "\u001b[36m(trainable_paddle_ocr pid=26140)\u001b[0m [2025-12-07 16:16:58,481 E 26140 16220] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\n", "\u001b[36m(trainable_paddle_ocr pid=20664)\u001b[0m [2025-12-07 16:17:16,506 E 20664 20720] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\n", "2025-12-07 16:22:27,297\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_111e5a9e_6_text_det_box_thresh=0.4833,text_det_thresh=0.0448,text_det_unclip_ratio=0.0000,text_rec_score_thre_2025-12-07_16-16-41\n", "2025-12-07 16:22:27,312\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_415d7ba1_7_text_det_box_thresh=0.5234,text_det_thresh=0.0170,text_det_unclip_ratio=0.0000,text_rec_score_thre_2025-12-07_16-22-27\n", "2025-12-07 16:22:27,316\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_415d7ba1_7_text_det_box_thresh=0.5234,text_det_thresh=0.0170,text_det_unclip_ratio=0.0000,text_rec_score_thre_2025-12-07_16-22-27\n", "2025-12-07 16:22:32,726\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_415d7ba1_7_text_det_box_thresh=0.5234,text_det_thresh=0.0170,text_det_unclip_ratio=0.0000,text_rec_score_thre_2025-12-07_16-22-27\n", "2025-12-07 16:22:32,728\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_415d7ba1_7_text_det_box_thresh=0.5234,text_det_thresh=0.0170,text_det_unclip_ratio=0.0000,text_rec_score_thre_2025-12-07_16-22-27\n", "2025-12-07 16:22:39,838\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_bf10d370_5_text_det_box_thresh=0.6902,text_det_thresh=0.6720,text_det_unclip_ratio=0.0000,text_rec_score_thre_2025-12-07_16-16-23\n", "2025-12-07 16:22:39,854\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_a58d8109_8_text_det_box_thresh=0.6706,text_det_thresh=0.0402,text_det_unclip_ratio=0.0000,text_rec_score_thre_2025-12-07_16-22-39\n", "2025-12-07 16:22:39,854\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_a58d8109_8_text_det_box_thresh=0.6706,text_det_thresh=0.0402,text_det_unclip_ratio=0.0000,text_rec_score_thre_2025-12-07_16-22-39\n", "2025-12-07 16:22:44,482\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_a58d8109_8_text_det_box_thresh=0.6706,text_det_thresh=0.0402,text_det_unclip_ratio=0.0000,text_rec_score_thre_2025-12-07_16-22-39\n", "2025-12-07 16:22:44,484\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_a58d8109_8_text_det_box_thresh=0.6706,text_det_thresh=0.0402,text_det_unclip_ratio=0.0000,text_rec_score_thre_2025-12-07_16-22-39\n", "\u001b[36m(trainable_paddle_ocr pid=23848)\u001b[0m [2025-12-07 16:23:02,571 E 23848 12908] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\n", "\u001b[36m(trainable_paddle_ocr pid=25248)\u001b[0m [2025-12-07 16:23:14,789 E 25248 4036] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\n", "2025-12-07 16:28:20,034\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_415d7ba1_7_text_det_box_thresh=0.5234,text_det_thresh=0.0170,text_det_unclip_ratio=0.0000,text_rec_score_thre_2025-12-07_16-22-27\n", "2025-12-07 16:28:20,052\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_33bdf2a9_9_text_det_box_thresh=0.4900,text_det_thresh=0.4347,text_det_unclip_ratio=0.0000,text_rec_score_thre_2025-12-07_16-28-20\n", "2025-12-07 16:28:20,055\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_33bdf2a9_9_text_det_box_thresh=0.4900,text_det_thresh=0.4347,text_det_unclip_ratio=0.0000,text_rec_score_thre_2025-12-07_16-28-20\n", "2025-12-07 16:28:24,790\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_33bdf2a9_9_text_det_box_thresh=0.4900,text_det_thresh=0.4347,text_det_unclip_ratio=0.0000,text_rec_score_thre_2025-12-07_16-28-20\n", "2025-12-07 16:28:24,790\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_33bdf2a9_9_text_det_box_thresh=0.4900,text_det_thresh=0.4347,text_det_unclip_ratio=0.0000,text_rec_score_thre_2025-12-07_16-28-20\n", "2025-12-07 16:28:30,585\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_a58d8109_8_text_det_box_thresh=0.6706,text_det_thresh=0.0402,text_det_unclip_ratio=0.0000,text_rec_score_thre_2025-12-07_16-22-39\n", "2025-12-07 16:28:30,605\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_d9df79f3_10_text_det_box_thresh=0.6262,text_det_thresh=0.1781,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_16-28-30\n", "2025-12-07 16:28:30,607\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_d9df79f3_10_text_det_box_thresh=0.6262,text_det_thresh=0.1781,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_16-28-30\n", "2025-12-07 16:28:35,143\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_d9df79f3_10_text_det_box_thresh=0.6262,text_det_thresh=0.1781,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_16-28-30\n", "2025-12-07 16:28:35,143\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_d9df79f3_10_text_det_box_thresh=0.6262,text_det_thresh=0.1781,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_16-28-30\n", "\u001b[36m(trainable_paddle_ocr pid=24024)\u001b[0m [2025-12-07 16:28:54,997 E 24024 23472] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\n", "\u001b[36m(trainable_paddle_ocr pid=5368)\u001b[0m [2025-12-07 16:29:05,433 E 5368 24544] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\n", "2025-12-07 16:34:52,986\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_33bdf2a9_9_text_det_box_thresh=0.4900,text_det_thresh=0.4347,text_det_unclip_ratio=0.0000,text_rec_score_thre_2025-12-07_16-28-20\n", "2025-12-07 16:34:53,020\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_80ea65f2_11_text_det_box_thresh=0.2514,text_det_thresh=0.6011,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_16-34-53\n", "2025-12-07 16:34:53,024\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_80ea65f2_11_text_det_box_thresh=0.2514,text_det_thresh=0.6011,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_16-34-53\n", "2025-12-07 16:34:58,668\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_80ea65f2_11_text_det_box_thresh=0.2514,text_det_thresh=0.6011,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_16-34-53\n", "2025-12-07 16:34:58,670\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_80ea65f2_11_text_det_box_thresh=0.2514,text_det_thresh=0.6011,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_16-34-53\n", "2025-12-07 16:34:59,856\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_d9df79f3_10_text_det_box_thresh=0.6262,text_det_thresh=0.1781,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_16-28-30\n", "2025-12-07 16:34:59,928\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_2e978bfa_12_text_det_box_thresh=0.0777,text_det_thresh=0.2349,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_16-34-59\n", "2025-12-07 16:34:59,933\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_2e978bfa_12_text_det_box_thresh=0.0777,text_det_thresh=0.2349,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_16-34-59\n", "2025-12-07 16:35:04,574\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_2e978bfa_12_text_det_box_thresh=0.0777,text_det_thresh=0.2349,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_16-34-59\n", "2025-12-07 16:35:04,576\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_2e978bfa_12_text_det_box_thresh=0.0777,text_det_thresh=0.2349,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_16-34-59\n", "\u001b[36m(trainable_paddle_ocr pid=14064)\u001b[0m [2025-12-07 16:35:28,312 E 14064 18904] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\n", "\u001b[36m(trainable_paddle_ocr pid=11060)\u001b[0m [2025-12-07 16:35:34,907 E 11060 16108] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\n", "2025-12-07 16:41:24,926\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_2e978bfa_12_text_det_box_thresh=0.0777,text_det_thresh=0.2349,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_16-34-59\n", "2025-12-07 16:41:24,993\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_8518cc40_13_text_det_box_thresh=0.0002,text_det_thresh=0.2226,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_16-41-24\n", "2025-12-07 16:41:24,996\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_8518cc40_13_text_det_box_thresh=0.0002,text_det_thresh=0.2226,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_16-41-24\n", "2025-12-07 16:41:26,379\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_80ea65f2_11_text_det_box_thresh=0.2514,text_det_thresh=0.6011,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_16-34-53\n", "2025-12-07 16:41:30,746\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_8518cc40_13_text_det_box_thresh=0.0002,text_det_thresh=0.2226,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_16-41-24\n", "2025-12-07 16:41:30,746\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_8518cc40_13_text_det_box_thresh=0.0002,text_det_thresh=0.2226,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_16-41-24\n", "2025-12-07 16:41:30,767\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_2c691aaa_14_text_det_box_thresh=0.0303,text_det_thresh=0.2247,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_16-41-30\n", "2025-12-07 16:41:30,770\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_2c691aaa_14_text_det_box_thresh=0.0303,text_det_thresh=0.2247,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_16-41-30\n", "2025-12-07 16:41:35,236\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_2c691aaa_14_text_det_box_thresh=0.0303,text_det_thresh=0.2247,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_16-41-30\n", "2025-12-07 16:41:35,236\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_2c691aaa_14_text_det_box_thresh=0.0303,text_det_thresh=0.2247,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_16-41-30\n", "\u001b[36m(trainable_paddle_ocr pid=21016)\u001b[0m [2025-12-07 16:42:00,269 E 21016 19044] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\n", "\u001b[36m(trainable_paddle_ocr pid=21540)\u001b[0m [2025-12-07 16:42:06,593 E 21540 1744] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\n", "2025-12-07 16:47:39,341\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_8518cc40_13_text_det_box_thresh=0.0002,text_det_thresh=0.2226,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_16-41-24\n", "2025-12-07 16:47:39,378\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_31e60691_15_text_det_box_thresh=0.0020,text_det_thresh=0.2591,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_16-47-39\n", "2025-12-07 16:47:39,378\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_31e60691_15_text_det_box_thresh=0.0020,text_det_thresh=0.2591,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_16-47-39\n", "2025-12-07 16:47:41,612\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_2c691aaa_14_text_det_box_thresh=0.0303,text_det_thresh=0.2247,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_16-41-30\n", "2025-12-07 16:47:44,526\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_31e60691_15_text_det_box_thresh=0.0020,text_det_thresh=0.2591,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_16-47-39\n", "2025-12-07 16:47:44,526\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_31e60691_15_text_det_box_thresh=0.0020,text_det_thresh=0.2591,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_16-47-39\n", "2025-12-07 16:47:44,541\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_d4d288c6_16_text_det_box_thresh=0.0034,text_det_thresh=0.2734,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_16-47-44\n", "2025-12-07 16:47:44,544\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_d4d288c6_16_text_det_box_thresh=0.0034,text_det_thresh=0.2734,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_16-47-44\n", "2025-12-07 16:47:49,055\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_d4d288c6_16_text_det_box_thresh=0.0034,text_det_thresh=0.2734,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_16-47-44\n", "2025-12-07 16:47:49,057\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_d4d288c6_16_text_det_box_thresh=0.0034,text_det_thresh=0.2734,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_16-47-44\n", "\u001b[36m(trainable_paddle_ocr pid=17532)\u001b[0m [2025-12-07 16:48:14,498 E 17532 10276] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\n", "2025-12-07 16:53:52,583\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_31e60691_15_text_det_box_thresh=0.0020,text_det_thresh=0.2591,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_16-47-39\n", "2025-12-07 16:53:52,603\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_7645b77c_17_text_det_box_thresh=0.1138,text_det_thresh=0.2792,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_16-53-52\n", "2025-12-07 16:53:52,608\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_7645b77c_17_text_det_box_thresh=0.1138,text_det_thresh=0.2792,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_16-53-52\n", "2025-12-07 16:53:57,961\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_d4d288c6_16_text_det_box_thresh=0.0034,text_det_thresh=0.2734,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_16-47-44\n", "2025-12-07 16:53:57,971\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_7645b77c_17_text_det_box_thresh=0.1138,text_det_thresh=0.2792,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_16-53-52\n", "2025-12-07 16:53:57,971\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_7645b77c_17_text_det_box_thresh=0.1138,text_det_thresh=0.2792,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_16-53-52\n", "2025-12-07 16:53:57,993\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_3256ae36_18_text_det_box_thresh=0.1292,text_det_thresh=0.3099,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_16-53-57\n", "2025-12-07 16:53:57,996\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_3256ae36_18_text_det_box_thresh=0.1292,text_det_thresh=0.3099,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_16-53-57\n", "2025-12-07 16:54:02,522\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_3256ae36_18_text_det_box_thresh=0.1292,text_det_thresh=0.3099,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_16-53-57\n", "2025-12-07 16:54:02,522\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_3256ae36_18_text_det_box_thresh=0.1292,text_det_thresh=0.3099,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_16-53-57\n", "\u001b[36m(trainable_paddle_ocr pid=2272)\u001b[0m [2025-12-07 16:54:27,753 E 2272 2144] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\u001b[32m [repeated 2x across cluster]\u001b[0m\n", "\u001b[36m(trainable_paddle_ocr pid=6604)\u001b[0m [2025-12-07 16:54:32,853 E 6604 7428] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\n", "2025-12-07 17:00:05,436\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_7645b77c_17_text_det_box_thresh=0.1138,text_det_thresh=0.2792,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_16-53-52\n", "2025-12-07 17:00:05,471\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_b0dda58b_19_text_det_box_thresh=0.1178,text_det_thresh=0.3150,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-00-05\n", "2025-12-07 17:00:05,471\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_b0dda58b_19_text_det_box_thresh=0.1178,text_det_thresh=0.3150,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-00-05\n", "2025-12-07 17:00:08,537\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_3256ae36_18_text_det_box_thresh=0.1292,text_det_thresh=0.3099,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_16-53-57\n", "2025-12-07 17:00:11,016\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_b0dda58b_19_text_det_box_thresh=0.1178,text_det_thresh=0.3150,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-00-05\n", "2025-12-07 17:00:11,017\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_b0dda58b_19_text_det_box_thresh=0.1178,text_det_thresh=0.3150,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-00-05\n", "2025-12-07 17:00:11,026\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_e9d40333_20_text_det_box_thresh=0.1569,text_det_thresh=0.5303,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-00-11\n", "2025-12-07 17:00:11,034\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_e9d40333_20_text_det_box_thresh=0.1569,text_det_thresh=0.5303,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-00-11\n", "2025-12-07 17:00:15,508\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_e9d40333_20_text_det_box_thresh=0.1569,text_det_thresh=0.5303,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-00-11\n", "2025-12-07 17:00:15,509\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_e9d40333_20_text_det_box_thresh=0.1569,text_det_thresh=0.5303,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-00-11\n", "\u001b[36m(trainable_paddle_ocr pid=9732)\u001b[0m [2025-12-07 17:00:40,741 E 9732 14552] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\n", "\u001b[36m(trainable_paddle_ocr pid=23416)\u001b[0m [2025-12-07 17:00:45,836 E 23416 4196] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\n", "2025-12-07 17:06:15,896\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_b0dda58b_19_text_det_box_thresh=0.1178,text_det_thresh=0.3150,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-00-05\n", "2025-12-07 17:06:15,950\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_aa89fe7a_21_text_det_box_thresh=0.1621,text_det_thresh=0.5040,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-06-15\n", "2025-12-07 17:06:15,953\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_aa89fe7a_21_text_det_box_thresh=0.1621,text_det_thresh=0.5040,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-06-15\n", "2025-12-07 17:06:21,172\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_e9d40333_20_text_det_box_thresh=0.1569,text_det_thresh=0.5303,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-00-11\n", "2025-12-07 17:06:21,708\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_aa89fe7a_21_text_det_box_thresh=0.1621,text_det_thresh=0.5040,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-06-15\n", "2025-12-07 17:06:21,709\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_aa89fe7a_21_text_det_box_thresh=0.1621,text_det_thresh=0.5040,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-06-15\n", "2025-12-07 17:06:21,722\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_92c48d07_22_text_det_box_thresh=0.1864,text_det_thresh=0.3332,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-06-21\n", "2025-12-07 17:06:21,724\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_92c48d07_22_text_det_box_thresh=0.1864,text_det_thresh=0.3332,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-06-21\n", "2025-12-07 17:06:26,213\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_92c48d07_22_text_det_box_thresh=0.1864,text_det_thresh=0.3332,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-06-21\n", "2025-12-07 17:06:26,213\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_92c48d07_22_text_det_box_thresh=0.1864,text_det_thresh=0.3332,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-06-21\n", "\u001b[36m(trainable_paddle_ocr pid=16200)\u001b[0m [2025-12-07 17:06:51,279 E 16200 7620] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\n", "\u001b[36m(trainable_paddle_ocr pid=15432)\u001b[0m [2025-12-07 17:06:56,512 E 15432 12008] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\n", "2025-12-07 17:12:28,470\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_aa89fe7a_21_text_det_box_thresh=0.1621,text_det_thresh=0.5040,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-06-15\n", "2025-12-07 17:12:28,508\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_187790d7_23_text_det_box_thresh=0.2353,text_det_thresh=0.3373,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-12-28\n", "2025-12-07 17:12:28,513\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_187790d7_23_text_det_box_thresh=0.2353,text_det_thresh=0.3373,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-12-28\n", "2025-12-07 17:12:31,317\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_92c48d07_22_text_det_box_thresh=0.1864,text_det_thresh=0.3332,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-06-21\n", "2025-12-07 17:12:33,695\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_187790d7_23_text_det_box_thresh=0.2353,text_det_thresh=0.3373,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-12-28\n", "2025-12-07 17:12:33,695\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_187790d7_23_text_det_box_thresh=0.2353,text_det_thresh=0.3373,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-12-28\n", "2025-12-07 17:12:33,716\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_442a2439_24_text_det_box_thresh=0.2123,text_det_thresh=0.5098,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-12-33\n", "2025-12-07 17:12:33,718\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_442a2439_24_text_det_box_thresh=0.2123,text_det_thresh=0.5098,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-12-33\n", "2025-12-07 17:12:38,168\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_442a2439_24_text_det_box_thresh=0.2123,text_det_thresh=0.5098,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-12-33\n", "2025-12-07 17:12:38,168\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_442a2439_24_text_det_box_thresh=0.2123,text_det_thresh=0.5098,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-12-33\n", "\u001b[36m(trainable_paddle_ocr pid=24676)\u001b[0m [2025-12-07 17:13:03,575 E 24676 21816] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\n", "2025-12-07 17:18:38,200\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_187790d7_23_text_det_box_thresh=0.2353,text_det_thresh=0.3373,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-12-28\n", "2025-12-07 17:18:38,251\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_70862adc_25_text_det_box_thresh=0.2163,text_det_thresh=0.3964,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-18-38\n", "2025-12-07 17:18:38,254\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_70862adc_25_text_det_box_thresh=0.2163,text_det_thresh=0.3964,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-18-38\n", "2025-12-07 17:18:42,934\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_442a2439_24_text_det_box_thresh=0.2123,text_det_thresh=0.5098,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-12-33\n", "2025-12-07 17:18:43,890\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_70862adc_25_text_det_box_thresh=0.2163,text_det_thresh=0.3964,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-18-38\n", "2025-12-07 17:18:43,892\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_70862adc_25_text_det_box_thresh=0.2163,text_det_thresh=0.3964,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-18-38\n", "2025-12-07 17:18:43,903\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_e6821f34_26_text_det_box_thresh=0.2408,text_det_thresh=0.3669,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-18-43\n", "2025-12-07 17:18:43,904\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_e6821f34_26_text_det_box_thresh=0.2408,text_det_thresh=0.3669,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-18-43\n", "2025-12-07 17:18:48,373\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_e6821f34_26_text_det_box_thresh=0.2408,text_det_thresh=0.3669,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-18-43\n", "2025-12-07 17:18:48,373\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_e6821f34_26_text_det_box_thresh=0.2408,text_det_thresh=0.3669,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-18-43\n", "\u001b[36m(trainable_paddle_ocr pid=15412)\u001b[0m [2025-12-07 17:19:13,443 E 15412 9512] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\u001b[32m [repeated 2x across cluster]\u001b[0m\n", "\u001b[36m(trainable_paddle_ocr pid=26088)\u001b[0m [2025-12-07 17:19:18,671 E 26088 10400] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\n", "2025-12-07 17:24:49,882\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_70862adc_25_text_det_box_thresh=0.2163,text_det_thresh=0.3964,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-18-38\n", "2025-12-07 17:24:49,909\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_8b680875_27_text_det_box_thresh=0.3193,text_det_thresh=0.5312,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-24-49\n", "2025-12-07 17:24:49,911\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_8b680875_27_text_det_box_thresh=0.3193,text_det_thresh=0.5312,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-24-49\n", "2025-12-07 17:24:53,650\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_e6821f34_26_text_det_box_thresh=0.2408,text_det_thresh=0.3669,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-18-43\n", "2025-12-07 17:24:55,137\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_8b680875_27_text_det_box_thresh=0.3193,text_det_thresh=0.5312,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-24-49\n", "2025-12-07 17:24:55,137\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_8b680875_27_text_det_box_thresh=0.3193,text_det_thresh=0.5312,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-24-49\n", "2025-12-07 17:24:55,153\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_fc54867b_28_text_det_box_thresh=0.3043,text_det_thresh=0.5034,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-24-55\n", "2025-12-07 17:24:55,156\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_fc54867b_28_text_det_box_thresh=0.3043,text_det_thresh=0.5034,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-24-55\n", "2025-12-07 17:24:59,622\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_fc54867b_28_text_det_box_thresh=0.3043,text_det_thresh=0.5034,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-24-55\n", "2025-12-07 17:24:59,622\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_fc54867b_28_text_det_box_thresh=0.3043,text_det_thresh=0.5034,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-24-55\n", "\u001b[36m(trainable_paddle_ocr pid=1720)\u001b[0m [2025-12-07 17:25:25,047 E 1720 25468] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\n", "2025-12-07 17:31:02,389\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_8b680875_27_text_det_box_thresh=0.3193,text_det_thresh=0.5312,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-24-49\n", "2025-12-07 17:31:02,469\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_c32d0d5e_29_text_det_box_thresh=0.3985,text_det_thresh=0.1530,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-31-02\n", "2025-12-07 17:31:02,473\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_c32d0d5e_29_text_det_box_thresh=0.3985,text_det_thresh=0.1530,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-31-02\n", "2025-12-07 17:31:08,377\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_fc54867b_28_text_det_box_thresh=0.3043,text_det_thresh=0.5034,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-24-55\n", "2025-12-07 17:31:08,467\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_c32d0d5e_29_text_det_box_thresh=0.3985,text_det_thresh=0.1530,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-31-02\n", "2025-12-07 17:31:08,467\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_c32d0d5e_29_text_det_box_thresh=0.3985,text_det_thresh=0.1530,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-31-02\n", "2025-12-07 17:31:08,487\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_4762fbbb_30_text_det_box_thresh=0.4010,text_det_thresh=0.1334,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-31-08\n", "2025-12-07 17:31:08,489\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_4762fbbb_30_text_det_box_thresh=0.4010,text_det_thresh=0.1334,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-31-08\n", "2025-12-07 17:31:12,960\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_4762fbbb_30_text_det_box_thresh=0.4010,text_det_thresh=0.1334,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-31-08\n", "2025-12-07 17:31:12,962\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_4762fbbb_30_text_det_box_thresh=0.4010,text_det_thresh=0.1334,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-31-08\n", "\u001b[36m(trainable_paddle_ocr pid=25808)\u001b[0m [2025-12-07 17:31:37,810 E 25808 21612] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\u001b[32m [repeated 2x across cluster]\u001b[0m\n", "\u001b[36m(trainable_paddle_ocr pid=20760)\u001b[0m [2025-12-07 17:31:43,311 E 20760 9512] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\n", "2025-12-07 17:37:12,922\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_c32d0d5e_29_text_det_box_thresh=0.3985,text_det_thresh=0.1530,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-31-02\n", "2025-12-07 17:37:12,971\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_522ac97c_31_text_det_box_thresh=0.4028,text_det_thresh=0.4490,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-37-12\n", "2025-12-07 17:37:12,975\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_522ac97c_31_text_det_box_thresh=0.4028,text_det_thresh=0.4490,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-37-12\n", "2025-12-07 17:37:16,310\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_4762fbbb_30_text_det_box_thresh=0.4010,text_det_thresh=0.1334,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-31-08\n", "2025-12-07 17:37:18,530\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_522ac97c_31_text_det_box_thresh=0.4028,text_det_thresh=0.4490,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-37-12\n", "2025-12-07 17:37:18,538\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_522ac97c_31_text_det_box_thresh=0.4028,text_det_thresh=0.4490,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-37-12\n", "2025-12-07 17:37:18,551\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_5784f433_32_text_det_box_thresh=0.1928,text_det_thresh=0.4620,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-37-18\n", "2025-12-07 17:37:18,553\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_5784f433_32_text_det_box_thresh=0.1928,text_det_thresh=0.4620,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-37-18\n", "2025-12-07 17:37:23,024\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_5784f433_32_text_det_box_thresh=0.1928,text_det_thresh=0.4620,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-37-18\n", "2025-12-07 17:37:23,030\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_5784f433_32_text_det_box_thresh=0.1928,text_det_thresh=0.4620,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-37-18\n", "\u001b[36m(trainable_paddle_ocr pid=2372)\u001b[0m [2025-12-07 17:37:49,189 E 2372 11208] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\n", "2025-12-07 17:43:23,269\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_522ac97c_31_text_det_box_thresh=0.4028,text_det_thresh=0.4490,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-37-12\n", "2025-12-07 17:43:23,297\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_83af0528_33_text_det_box_thresh=0.1846,text_det_thresh=0.4663,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-43-23\n", "2025-12-07 17:43:23,299\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_83af0528_33_text_det_box_thresh=0.1846,text_det_thresh=0.4663,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-43-23\n", "2025-12-07 17:43:25,962\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_5784f433_32_text_det_box_thresh=0.1928,text_det_thresh=0.4620,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-37-18\n", "2025-12-07 17:43:28,377\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_83af0528_33_text_det_box_thresh=0.1846,text_det_thresh=0.4663,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-43-23\n", "2025-12-07 17:43:28,377\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_83af0528_33_text_det_box_thresh=0.1846,text_det_thresh=0.4663,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-43-23\n", "2025-12-07 17:43:28,392\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_12cbaa22_34_text_det_box_thresh=0.4056,text_det_thresh=0.4728,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-43-28\n", "2025-12-07 17:43:28,394\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_12cbaa22_34_text_det_box_thresh=0.4056,text_det_thresh=0.4728,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-43-28\n", "2025-12-07 17:43:32,822\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_12cbaa22_34_text_det_box_thresh=0.4056,text_det_thresh=0.4728,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-43-28\n", "2025-12-07 17:43:32,822\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_12cbaa22_34_text_det_box_thresh=0.4056,text_det_thresh=0.4728,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-43-28\n", "\u001b[36m(trainable_paddle_ocr pid=9832)\u001b[0m [2025-12-07 17:43:58,320 E 9832 20188] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\u001b[32m [repeated 2x across cluster]\u001b[0m\n", "2025-12-07 17:49:32,969\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_83af0528_33_text_det_box_thresh=0.1846,text_det_thresh=0.4663,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-43-23\n", "2025-12-07 17:49:32,999\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_a3a87765_35_text_det_box_thresh=0.2856,text_det_thresh=0.4501,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-49-32\n", "2025-12-07 17:49:33,002\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_a3a87765_35_text_det_box_thresh=0.2856,text_det_thresh=0.4501,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-49-32\n", "2025-12-07 17:49:37,086\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_12cbaa22_34_text_det_box_thresh=0.4056,text_det_thresh=0.4728,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-43-28\n", "2025-12-07 17:49:38,207\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_a3a87765_35_text_det_box_thresh=0.2856,text_det_thresh=0.4501,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-49-32\n", "2025-12-07 17:49:38,207\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_a3a87765_35_text_det_box_thresh=0.2856,text_det_thresh=0.4501,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-49-32\n", "2025-12-07 17:49:38,221\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_cf2bad0c_36_text_det_box_thresh=0.2837,text_det_thresh=0.5890,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-49-38\n", "2025-12-07 17:49:38,224\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_cf2bad0c_36_text_det_box_thresh=0.2837,text_det_thresh=0.5890,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-49-38\n", "2025-12-07 17:49:42,732\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_cf2bad0c_36_text_det_box_thresh=0.2837,text_det_thresh=0.5890,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-49-38\n", "2025-12-07 17:49:42,734\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_cf2bad0c_36_text_det_box_thresh=0.2837,text_det_thresh=0.5890,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-49-38\n", "\u001b[36m(trainable_paddle_ocr pid=24372)\u001b[0m [2025-12-07 17:50:08,047 E 24372 25404] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\u001b[32m [repeated 2x across cluster]\u001b[0m\n", "\u001b[36m(trainable_paddle_ocr pid=3272)\u001b[0m [2025-12-07 17:50:14,041 E 3272 25236] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\n", "2025-12-07 17:55:47,492\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_a3a87765_35_text_det_box_thresh=0.2856,text_det_thresh=0.4501,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-49-32\n", "2025-12-07 17:55:47,513\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_9a9b91e7_37_text_det_box_thresh=0.3646,text_det_thresh=0.6090,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-55-47\n", "2025-12-07 17:55:47,515\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_9a9b91e7_37_text_det_box_thresh=0.3646,text_det_thresh=0.6090,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-55-47\n", "2025-12-07 17:55:48,925\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_cf2bad0c_36_text_det_box_thresh=0.2837,text_det_thresh=0.5890,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-49-38\n", "2025-12-07 17:55:52,512\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_9a9b91e7_37_text_det_box_thresh=0.3646,text_det_thresh=0.6090,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-55-47\n", "2025-12-07 17:55:52,520\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_9a9b91e7_37_text_det_box_thresh=0.3646,text_det_thresh=0.6090,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-55-47\n", "2025-12-07 17:55:52,532\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_e326d901_38_text_det_box_thresh=0.3735,text_det_thresh=0.5932,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-55-52\n", "2025-12-07 17:55:52,532\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_e326d901_38_text_det_box_thresh=0.3735,text_det_thresh=0.5932,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-55-52\n", "2025-12-07 17:55:56,990\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_e326d901_38_text_det_box_thresh=0.3735,text_det_thresh=0.5932,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-55-52\n", "2025-12-07 17:55:56,990\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_e326d901_38_text_det_box_thresh=0.3735,text_det_thresh=0.5932,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-55-52\n", "\u001b[36m(trainable_paddle_ocr pid=2272)\u001b[0m [2025-12-07 17:56:22,469 E 2272 9344] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\n", "2025-12-07 18:01:56,576\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_9a9b91e7_37_text_det_box_thresh=0.3646,text_det_thresh=0.6090,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-55-47\n", "2025-12-07 18:01:56,635\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_ccb3f19a_39_text_det_box_thresh=0.4538,text_det_thresh=0.6866,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-01-56\n", "2025-12-07 18:01:56,637\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_ccb3f19a_39_text_det_box_thresh=0.4538,text_det_thresh=0.6866,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-01-56\n", "2025-12-07 18:02:02,426\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_ccb3f19a_39_text_det_box_thresh=0.4538,text_det_thresh=0.6866,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-01-56\n", "2025-12-07 18:02:02,426\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_ccb3f19a_39_text_det_box_thresh=0.4538,text_det_thresh=0.6866,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-01-56\n", "2025-12-07 18:02:02,442\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_e326d901_38_text_det_box_thresh=0.3735,text_det_thresh=0.5932,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_17-55-52\n", "2025-12-07 18:02:02,471\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_8c12c55f_40_text_det_box_thresh=0.4444,text_det_thresh=0.6710,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-02-02\n", "2025-12-07 18:02:02,472\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_8c12c55f_40_text_det_box_thresh=0.4444,text_det_thresh=0.6710,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-02-02\n", "2025-12-07 18:02:06,950\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_8c12c55f_40_text_det_box_thresh=0.4444,text_det_thresh=0.6710,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-02-02\n", "2025-12-07 18:02:06,950\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_8c12c55f_40_text_det_box_thresh=0.4444,text_det_thresh=0.6710,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-02-02\n", "\u001b[36m(trainable_paddle_ocr pid=1104)\u001b[0m [2025-12-07 18:02:31,870 E 1104 11720] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\u001b[32m [repeated 2x across cluster]\u001b[0m\n", "\u001b[36m(trainable_paddle_ocr pid=19700)\u001b[0m [2025-12-07 18:02:38,333 E 19700 6824] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\n", "2025-12-07 18:08:07,593\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_ccb3f19a_39_text_det_box_thresh=0.4538,text_det_thresh=0.6866,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-01-56\n", "2025-12-07 18:08:07,628\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_5a62d5b6_41_text_det_box_thresh=0.2010,text_det_thresh=0.4041,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-08-07\n", "2025-12-07 18:08:07,630\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_5a62d5b6_41_text_det_box_thresh=0.2010,text_det_thresh=0.4041,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-08-07\n", "2025-12-07 18:08:10,260\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_8c12c55f_40_text_det_box_thresh=0.4444,text_det_thresh=0.6710,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-02-02\n", "2025-12-07 18:08:12,660\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_5a62d5b6_41_text_det_box_thresh=0.2010,text_det_thresh=0.4041,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-08-07\n", "2025-12-07 18:08:12,664\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_5a62d5b6_41_text_det_box_thresh=0.2010,text_det_thresh=0.4041,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-08-07\n", "2025-12-07 18:08:12,675\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_bb4495b7_42_text_det_box_thresh=0.5764,text_det_thresh=0.3907,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-08-12\n", "2025-12-07 18:08:12,684\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_bb4495b7_42_text_det_box_thresh=0.5764,text_det_thresh=0.3907,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-08-12\n", "2025-12-07 18:08:17,160\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_bb4495b7_42_text_det_box_thresh=0.5764,text_det_thresh=0.3907,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-08-12\n", "2025-12-07 18:08:17,164\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_bb4495b7_42_text_det_box_thresh=0.5764,text_det_thresh=0.3907,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-08-12\n", "\u001b[36m(trainable_paddle_ocr pid=26528)\u001b[0m [2025-12-07 18:08:42,646 E 26528 5412] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\n", "\u001b[36m(trainable_paddle_ocr pid=21772)\u001b[0m [2025-12-07 18:08:48,607 E 21772 12564] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\n", "2025-12-07 18:14:33,027\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_5a62d5b6_41_text_det_box_thresh=0.2010,text_det_thresh=0.4041,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-08-07\n", "2025-12-07 18:14:33,082\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_9d90711d_43_text_det_box_thresh=0.5412,text_det_thresh=0.4690,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-14-33\n", "2025-12-07 18:14:33,085\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_9d90711d_43_text_det_box_thresh=0.5412,text_det_thresh=0.4690,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-14-33\n", "2025-12-07 18:14:33,144\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_bb4495b7_42_text_det_box_thresh=0.5764,text_det_thresh=0.3907,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-08-12\n", "2025-12-07 18:14:38,712\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_9d90711d_43_text_det_box_thresh=0.5412,text_det_thresh=0.4690,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-14-33\n", "2025-12-07 18:14:38,714\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_9d90711d_43_text_det_box_thresh=0.5412,text_det_thresh=0.4690,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-14-33\n", "2025-12-07 18:14:38,727\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_daaec3f8_44_text_det_box_thresh=0.5213,text_det_thresh=0.4744,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-14-38\n", "2025-12-07 18:14:38,731\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_daaec3f8_44_text_det_box_thresh=0.5213,text_det_thresh=0.4744,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-14-38\n", "2025-12-07 18:14:43,202\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_daaec3f8_44_text_det_box_thresh=0.5213,text_det_thresh=0.4744,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-14-38\n", "2025-12-07 18:14:43,206\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_daaec3f8_44_text_det_box_thresh=0.5213,text_det_thresh=0.4744,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-14-38\n", "\u001b[36m(trainable_paddle_ocr pid=17592)\u001b[0m [2025-12-07 18:15:08,237 E 17592 11980] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\n", "\u001b[36m(trainable_paddle_ocr pid=21292)\u001b[0m [2025-12-07 18:15:13,513 E 21292 10368] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\n", "2025-12-07 18:20:44,494\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_9d90711d_43_text_det_box_thresh=0.5412,text_det_thresh=0.4690,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-14-33\n", "2025-12-07 18:20:44,525\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_51fb5915_45_text_det_box_thresh=0.5811,text_det_thresh=0.4854,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-20-44\n", "2025-12-07 18:20:44,528\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_51fb5915_45_text_det_box_thresh=0.5811,text_det_thresh=0.4854,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-20-44\n", "2025-12-07 18:20:46,235\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_daaec3f8_44_text_det_box_thresh=0.5213,text_det_thresh=0.4744,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-14-38\n", "2025-12-07 18:20:49,638\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_51fb5915_45_text_det_box_thresh=0.5811,text_det_thresh=0.4854,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-20-44\n", "2025-12-07 18:20:49,639\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_51fb5915_45_text_det_box_thresh=0.5811,text_det_thresh=0.4854,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-20-44\n", "2025-12-07 18:20:49,649\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_18966a33_46_text_det_box_thresh=0.5133,text_det_thresh=0.5502,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-20-49\n", "2025-12-07 18:20:49,649\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_18966a33_46_text_det_box_thresh=0.5133,text_det_thresh=0.5502,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-20-49\n", "2025-12-07 18:20:54,162\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_18966a33_46_text_det_box_thresh=0.5133,text_det_thresh=0.5502,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-20-49\n", "2025-12-07 18:20:54,162\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_18966a33_46_text_det_box_thresh=0.5133,text_det_thresh=0.5502,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-20-49\n", "\u001b[36m(trainable_paddle_ocr pid=21772)\u001b[0m [2025-12-07 18:21:19,532 E 21772 9096] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\n", "2025-12-07 18:26:53,700\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_51fb5915_45_text_det_box_thresh=0.5811,text_det_thresh=0.4854,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-20-44\n", "2025-12-07 18:26:53,763\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_b67080f9_47_text_det_box_thresh=0.5761,text_det_thresh=0.5534,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-26-53\n", "2025-12-07 18:26:53,766\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_b67080f9_47_text_det_box_thresh=0.5761,text_det_thresh=0.5534,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-26-53\n", "2025-12-07 18:26:57,513\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_18966a33_46_text_det_box_thresh=0.5133,text_det_thresh=0.5502,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-20-49\n", "2025-12-07 18:26:59,363\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_b67080f9_47_text_det_box_thresh=0.5761,text_det_thresh=0.5534,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-26-53\n", "2025-12-07 18:26:59,363\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_b67080f9_47_text_det_box_thresh=0.5761,text_det_thresh=0.5534,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-26-53\n", "2025-12-07 18:26:59,379\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_2533f368_48_text_det_box_thresh=0.5246,text_det_thresh=0.5572,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-26-59\n", "2025-12-07 18:26:59,382\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_2533f368_48_text_det_box_thresh=0.5246,text_det_thresh=0.5572,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-26-59\n", "2025-12-07 18:27:03,913\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_2533f368_48_text_det_box_thresh=0.5246,text_det_thresh=0.5572,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-26-59\n", "2025-12-07 18:27:03,913\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_2533f368_48_text_det_box_thresh=0.5246,text_det_thresh=0.5572,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-26-59\n", "\u001b[36m(trainable_paddle_ocr pid=20948)\u001b[0m [2025-12-07 18:27:29,044 E 20948 19656] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\u001b[32m [repeated 2x across cluster]\u001b[0m\n", "\u001b[36m(trainable_paddle_ocr pid=11208)\u001b[0m [2025-12-07 18:27:34,203 E 11208 2320] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\n", "2025-12-07 18:33:05,400\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_b67080f9_47_text_det_box_thresh=0.5761,text_det_thresh=0.5534,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-26-53\n", "2025-12-07 18:33:05,427\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_451d018d_49_text_det_box_thresh=0.5495,text_det_thresh=0.6340,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-33-05\n", "2025-12-07 18:33:05,428\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_451d018d_49_text_det_box_thresh=0.5495,text_det_thresh=0.6340,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-33-05\n", "2025-12-07 18:33:10,740\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_451d018d_49_text_det_box_thresh=0.5495,text_det_thresh=0.6340,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-33-05\n", "2025-12-07 18:33:10,743\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_451d018d_49_text_det_box_thresh=0.5495,text_det_thresh=0.6340,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-33-05\n", "2025-12-07 18:33:15,130\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_2533f368_48_text_det_box_thresh=0.5246,text_det_thresh=0.5572,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-26-59\n", "2025-12-07 18:33:15,154\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_2256e752_50_text_det_box_thresh=0.6229,text_det_thresh=0.6478,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-33-15\n", "2025-12-07 18:33:15,156\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_2256e752_50_text_det_box_thresh=0.6229,text_det_thresh=0.6478,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-33-15\n", "2025-12-07 18:33:19,685\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_2256e752_50_text_det_box_thresh=0.6229,text_det_thresh=0.6478,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-33-15\n", "2025-12-07 18:33:19,685\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_2256e752_50_text_det_box_thresh=0.6229,text_det_thresh=0.6478,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-33-15\n", "\u001b[36m(trainable_paddle_ocr pid=3616)\u001b[0m [2025-12-07 18:33:40,534 E 3616 22824] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\n", "\u001b[36m(trainable_paddle_ocr pid=25468)\u001b[0m [2025-12-07 18:33:49,934 E 25468 7192] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\n", "2025-12-07 18:39:29,627\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_2256e752_50_text_det_box_thresh=0.6229,text_det_thresh=0.6478,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-33-15\n", "2025-12-07 18:39:29,649\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_451d018d_49_text_det_box_thresh=0.5495,text_det_thresh=0.6340,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-33-05\n", "2025-12-07 18:39:29,687\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_0a892729_51_text_det_box_thresh=0.5429,text_det_thresh=0.4217,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-39-29\n", "2025-12-07 18:39:29,690\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_0a892729_51_text_det_box_thresh=0.5429,text_det_thresh=0.4217,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-39-29\n", "2025-12-07 18:39:35,040\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_0a892729_51_text_det_box_thresh=0.5429,text_det_thresh=0.4217,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-39-29\n", "2025-12-07 18:39:35,040\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_0a892729_51_text_det_box_thresh=0.5429,text_det_thresh=0.4217,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-39-29\n", "2025-12-07 18:39:35,057\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_495075f5_52_text_det_box_thresh=0.6319,text_det_thresh=0.4187,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-39-35\n", "2025-12-07 18:39:35,059\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_495075f5_52_text_det_box_thresh=0.6319,text_det_thresh=0.4187,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-39-35\n", "2025-12-07 18:39:39,597\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_495075f5_52_text_det_box_thresh=0.6319,text_det_thresh=0.4187,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-39-35\n", "2025-12-07 18:39:39,598\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_495075f5_52_text_det_box_thresh=0.6319,text_det_thresh=0.4187,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-39-35\n", "\u001b[36m(trainable_paddle_ocr pid=26212)\u001b[0m [2025-12-07 18:40:04,811 E 26212 22100] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\n", "\u001b[36m(trainable_paddle_ocr pid=23604)\u001b[0m [2025-12-07 18:40:10,081 E 23604 16924] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\n", "2025-12-07 18:45:42,301\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_0a892729_51_text_det_box_thresh=0.5429,text_det_thresh=0.4217,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-39-29\n", "2025-12-07 18:45:42,331\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_54c45552_53_text_det_box_thresh=0.6197,text_det_thresh=0.4638,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-45-42\n", "2025-12-07 18:45:42,335\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_54c45552_53_text_det_box_thresh=0.6197,text_det_thresh=0.4638,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-45-42\n", "2025-12-07 18:45:45,144\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_495075f5_52_text_det_box_thresh=0.6319,text_det_thresh=0.4187,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-39-35\n", "2025-12-07 18:45:47,422\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_54c45552_53_text_det_box_thresh=0.6197,text_det_thresh=0.4638,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-45-42\n", "2025-12-07 18:45:47,422\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_54c45552_53_text_det_box_thresh=0.6197,text_det_thresh=0.4638,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-45-42\n", "2025-12-07 18:45:47,436\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_6b2e9b93_54_text_det_box_thresh=0.4893,text_det_thresh=0.4752,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-45-47\n", "2025-12-07 18:45:47,436\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_6b2e9b93_54_text_det_box_thresh=0.4893,text_det_thresh=0.4752,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-45-47\n", "2025-12-07 18:45:51,980\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_6b2e9b93_54_text_det_box_thresh=0.4893,text_det_thresh=0.4752,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-45-47\n", "2025-12-07 18:45:51,980\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_6b2e9b93_54_text_det_box_thresh=0.4893,text_det_thresh=0.4752,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-45-47\n", "\u001b[36m(trainable_paddle_ocr pid=25352)\u001b[0m [2025-12-07 18:46:17,386 E 25352 26068] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\n", "2025-12-07 18:51:55,425\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_54c45552_53_text_det_box_thresh=0.6197,text_det_thresh=0.4638,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-45-42\n", "2025-12-07 18:51:55,497\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_e9a6b81f_55_text_det_box_thresh=0.4926,text_det_thresh=0.4879,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-51-55\n", "2025-12-07 18:51:55,501\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_e9a6b81f_55_text_det_box_thresh=0.4926,text_det_thresh=0.4879,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-51-55\n", "2025-12-07 18:51:57,995\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_6b2e9b93_54_text_det_box_thresh=0.4893,text_det_thresh=0.4752,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-45-47\n", "2025-12-07 18:52:01,238\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_e9a6b81f_55_text_det_box_thresh=0.4926,text_det_thresh=0.4879,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-51-55\n", "2025-12-07 18:52:01,239\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_e9a6b81f_55_text_det_box_thresh=0.4926,text_det_thresh=0.4879,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-51-55\n", "2025-12-07 18:52:01,255\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_076c5450_56_text_det_box_thresh=0.5881,text_det_thresh=0.4884,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-52-01\n", "2025-12-07 18:52:01,258\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_076c5450_56_text_det_box_thresh=0.5881,text_det_thresh=0.4884,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-52-01\n", "2025-12-07 18:52:05,685\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_076c5450_56_text_det_box_thresh=0.5881,text_det_thresh=0.4884,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-52-01\n", "2025-12-07 18:52:05,685\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_076c5450_56_text_det_box_thresh=0.5881,text_det_thresh=0.4884,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-52-01\n", "\u001b[36m(trainable_paddle_ocr pid=4036)\u001b[0m [2025-12-07 18:52:30,776 E 4036 16404] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\u001b[32m [repeated 2x across cluster]\u001b[0m\n", "\u001b[36m(trainable_paddle_ocr pid=4832)\u001b[0m [2025-12-07 18:52:36,982 E 4832 22740] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\n", "2025-12-07 18:58:08,591\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_e9a6b81f_55_text_det_box_thresh=0.4926,text_det_thresh=0.4879,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-51-55\n", "2025-12-07 18:58:08,621\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_4a42a3ea_57_text_det_box_thresh=0.5940,text_det_thresh=0.5590,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-58-08\n", "2025-12-07 18:58:08,624\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_4a42a3ea_57_text_det_box_thresh=0.5940,text_det_thresh=0.5590,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-58-08\n", "2025-12-07 18:58:10,886\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_076c5450_56_text_det_box_thresh=0.5881,text_det_thresh=0.4884,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-52-01\n", "2025-12-07 18:58:13,816\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_4a42a3ea_57_text_det_box_thresh=0.5940,text_det_thresh=0.5590,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-58-08\n", "2025-12-07 18:58:13,816\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_4a42a3ea_57_text_det_box_thresh=0.5940,text_det_thresh=0.5590,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-58-08\n", "2025-12-07 18:58:13,830\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_041795f1_58_text_det_box_thresh=0.6617,text_det_thresh=0.5650,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-58-13\n", "2025-12-07 18:58:13,833\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_041795f1_58_text_det_box_thresh=0.6617,text_det_thresh=0.5650,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-58-13\n", "2025-12-07 18:58:18,273\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_041795f1_58_text_det_box_thresh=0.6617,text_det_thresh=0.5650,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-58-13\n", "2025-12-07 18:58:18,280\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_041795f1_58_text_det_box_thresh=0.6617,text_det_thresh=0.5650,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-58-13\n", "\u001b[36m(trainable_paddle_ocr pid=14912)\u001b[0m [2025-12-07 18:58:43,671 E 14912 9648] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\n", "2025-12-07 19:04:24,842\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_4a42a3ea_57_text_det_box_thresh=0.5940,text_det_thresh=0.5590,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-58-08\n", "2025-12-07 19:04:24,907\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_8abb3f37_59_text_det_box_thresh=0.4637,text_det_thresh=0.4898,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_19-04-24\n", "2025-12-07 19:04:24,910\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_8abb3f37_59_text_det_box_thresh=0.4637,text_det_thresh=0.4898,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_19-04-24\n", "2025-12-07 19:04:29,252\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_041795f1_58_text_det_box_thresh=0.6617,text_det_thresh=0.5650,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_18-58-13\n", "2025-12-07 19:04:30,602\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_8abb3f37_59_text_det_box_thresh=0.4637,text_det_thresh=0.4898,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_19-04-24\n", "2025-12-07 19:04:30,603\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_8abb3f37_59_text_det_box_thresh=0.4637,text_det_thresh=0.4898,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_19-04-24\n", "2025-12-07 19:04:30,613\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_f2cb682e_60_text_det_box_thresh=0.4522,text_det_thresh=0.4918,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_19-04-30\n", "2025-12-07 19:04:30,619\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_f2cb682e_60_text_det_box_thresh=0.4522,text_det_thresh=0.4918,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_19-04-30\n", "2025-12-07 19:04:35,119\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_f2cb682e_60_text_det_box_thresh=0.4522,text_det_thresh=0.4918,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_19-04-30\n", "2025-12-07 19:04:35,119\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_f2cb682e_60_text_det_box_thresh=0.4522,text_det_thresh=0.4918,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_19-04-30\n", "\u001b[36m(trainable_paddle_ocr pid=22012)\u001b[0m [2025-12-07 19:05:01,269 E 22012 4372] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\u001b[32m [repeated 2x across cluster]\u001b[0m\n", "2025-12-07 19:10:35,351\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_8abb3f37_59_text_det_box_thresh=0.4637,text_det_thresh=0.4898,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_19-04-24\n", "2025-12-07 19:10:35,442\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_463fe5e7_61_text_det_box_thresh=0.5202,text_det_thresh=0.5373,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_19-10-35\n", "2025-12-07 19:10:35,445\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_463fe5e7_61_text_det_box_thresh=0.5202,text_det_thresh=0.5373,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_19-10-35\n", "2025-12-07 19:10:40,065\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_f2cb682e_60_text_det_box_thresh=0.4522,text_det_thresh=0.4918,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_19-04-30\n", "2025-12-07 19:10:41,249\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_463fe5e7_61_text_det_box_thresh=0.5202,text_det_thresh=0.5373,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_19-10-35\n", "2025-12-07 19:10:41,249\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_463fe5e7_61_text_det_box_thresh=0.5202,text_det_thresh=0.5373,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_19-10-35\n", "2025-12-07 19:10:41,261\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_88bbe87d_62_text_det_box_thresh=0.5111,text_det_thresh=0.5275,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_19-10-41\n", "2025-12-07 19:10:41,261\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_88bbe87d_62_text_det_box_thresh=0.5111,text_det_thresh=0.5275,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_19-10-41\n", "2025-12-07 19:10:45,749\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_88bbe87d_62_text_det_box_thresh=0.5111,text_det_thresh=0.5275,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_19-10-41\n", "2025-12-07 19:10:45,750\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_88bbe87d_62_text_det_box_thresh=0.5111,text_det_thresh=0.5275,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_19-10-41\n", "\u001b[36m(trainable_paddle_ocr pid=16524)\u001b[0m [2025-12-07 19:11:10,747 E 16524 6148] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\u001b[32m [repeated 2x across cluster]\u001b[0m\n", "\u001b[36m(trainable_paddle_ocr pid=15084)\u001b[0m [2025-12-07 19:11:16,039 E 15084 20216] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\n", "2025-12-07 19:16:51,841\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_463fe5e7_61_text_det_box_thresh=0.5202,text_det_thresh=0.5373,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_19-10-35\n", "2025-12-07 19:16:51,883\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_33ea1cc6_63_text_det_box_thresh=0.5158,text_det_thresh=0.5230,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_19-16-51\n", "2025-12-07 19:16:51,884\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_33ea1cc6_63_text_det_box_thresh=0.5158,text_det_thresh=0.5230,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_19-16-51\n", "2025-12-07 19:16:55,313\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_88bbe87d_62_text_det_box_thresh=0.5111,text_det_thresh=0.5275,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_19-10-41\n", "2025-12-07 19:16:57,623\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_33ea1cc6_63_text_det_box_thresh=0.5158,text_det_thresh=0.5230,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_19-16-51\n", "2025-12-07 19:16:57,623\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_33ea1cc6_63_text_det_box_thresh=0.5158,text_det_thresh=0.5230,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_19-16-51\n", "2025-12-07 19:16:57,638\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_1243723e_64_text_det_box_thresh=0.5573,text_det_thresh=0.3727,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_19-16-57\n", "2025-12-07 19:16:57,639\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_1243723e_64_text_det_box_thresh=0.5573,text_det_thresh=0.3727,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_19-16-57\n", "2025-12-07 19:17:02,358\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_1243723e_64_text_det_box_thresh=0.5573,text_det_thresh=0.3727,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_19-16-57\n", "2025-12-07 19:17:02,362\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_1243723e_64_text_det_box_thresh=0.5573,text_det_thresh=0.3727,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_19-16-57\n", "\u001b[36m(trainable_paddle_ocr pid=17380)\u001b[0m [2025-12-07 19:17:27,300 E 17380 17224] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\n", "\u001b[36m(trainable_paddle_ocr pid=11232)\u001b[0m [2025-12-07 19:17:32,685 E 11232 7916] core_worker_process.cc:837: Failed to establish connection to the metrics exporter agent. Metrics will not be exported. Exporter agent status: RpcError: Running out of retries to initialize the metrics agent. rpc_code: 14\n", "2025-12-07 19:23:14,420\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_33ea1cc6_63_text_det_box_thresh=0.5158,text_det_thresh=0.5230,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_19-16-51\n", "2025-12-07 19:23:17,826\tWARNING trial.py:647 -- The path to the trial log directory is too long (max length: 260. Consider using `trial_dirname_creator` to shorten the path. Path: C:\\Users\\Sergio\\AppData\\Local\\Temp\\ray\\session_2025-12-07_15-57-58_291425_24012\\artifacts\\2025-12-07_16-03-56\\trainable_paddle_ocr_2025-12-07_16-03-56\\driver_artifacts\\trainable_paddle_ocr_1243723e_64_text_det_box_thresh=0.5573,text_det_thresh=0.3727,text_det_unclip_ratio=0.0000,text_rec_score_thr_2025-12-07_19-16-57\n", "2025-12-07 19:23:17,928\tINFO tune.py:1009 -- Wrote the latest version of all result files and experiment state to 'C:/Users/Sergio/ray_results/trainable_paddle_ocr_2025-12-07_16-03-56' in 0.0859s.\n", "2025-12-07 19:23:17,957\tINFO tune.py:1041 -- Total run time: 11961.30 seconds (11961.14 seconds for the tuning loop).\n" ] } ], "source": [ "from ray.tune.search.optuna import OptunaSearch\n", "\n", "def trainable_paddle_ocr(config):\n", " args = [sys.executable, SCRIPT_ABS, \"--pdf-folder\", PDF_FOLDER_ABS]\n", " for k, v in config.items():\n", " args += [f\"--{KEYMAP[k]}\", str(v)]\n", " proc = subprocess.run(args, capture_output=True, text=True, cwd=SCRIPT_DIR)\n", "\n", " if proc.returncode != 0:\n", " tune.report({\"CER\": 1.0, \"WER\": 1.0, \"TIME\": 0.0, 'PAGES': 0, 'TIME_PER_PAGE': 0, \"ERROR\": proc.stderr[:500]})\n", " return\n", " # last line contains the metrics in json format\n", " last = proc.stdout.strip().splitlines()[-1]\n", " \n", " metrics = json.loads(last)\n", " tune.report(metrics=metrics)\n", "\n", "tuner = tune.Tuner(\n", " trainable_paddle_ocr,\n", " tune_config=tune.TuneConfig(metric=\"CER\", \n", " mode=\"min\", \n", " search_alg=OptunaSearch(),\n", " num_samples=64, \n", " max_concurrent_trials=2),\n", " run_config=air.RunConfig(verbose=2, log_to_file=False),\n", " param_space=search_space\n", ")\n", "\n", "results = tuner.fit()" ] }, { "cell_type": "code", "execution_count": null, "id": "710a67ce", "metadata": {}, "outputs": [], "source": [ "df = results.get_dataframe()" ] }, { "cell_type": "code", "execution_count": null, "id": "1ab345a3", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Guardado: raytune_paddle_subproc_results_20251207_192320.csv\n" ] } ], "source": [ "# Generate a unique filename with timestamp\n", "timestamp = datetime.now().strftime(\"%Y%m%d_%H%M%S\")\n", "filename = f\"raytune_paddle_subproc_results_{timestamp}.csv\"\n", "filepath = os.path.join(OUTPUT_FOLDER, filename)\n", "\n", "\n", "df.to_csv(filename, index=False)\n", "print(f\"Guardado: {filename}\")" ] }, { "cell_type": "code", "execution_count": null, "id": "3e3a34e4", "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
CERWERTIMEPAGESTIME_PER_PAGEtimestampcheckpoint_dir_nametraining_iterationtime_this_iter_stime_total_spidtime_since_restoreiterations_since_restoreconfig/text_det_threshconfig/text_det_box_threshconfig/text_det_unclip_ratioconfig/text_rec_score_thresh
count64.00000064.00000064.00000064.064.0000006.400000e+010.064.064.00000064.00000064.00000064.00000064.064.00000064.00000064.064.000000
mean0.0524820.142770347.6058705.069.4237341.765126e+09NaN1.0367.715945367.71594516306.750000367.7159451.00.4190910.3929650.00.470584
std0.1102690.1075157.8765390.01.5744703.473487e+03NaN0.08.0115548.0115548179.9171148.0115540.00.1671780.1954190.00.219216
min0.0115350.098902320.9662055.064.0952101.765120e+09NaN1.0341.071264341.0712641104.000000341.0712641.00.0169970.0002420.00.002891
25%0.0119680.100441344.2391165.068.7551181.765123e+09NaN1.0364.708660364.7086609272.000000364.7086601.00.3286520.2305150.00.311325
50%0.0123140.102033346.4196825.069.1888751.765126e+09NaN1.0366.103412366.10341218522.000000366.1034121.00.4650680.4483320.00.559640
75%0.0403390.132047350.1445635.069.9301731.765129e+09NaN1.0370.648662370.64866223167.000000370.6486621.00.5305010.5445630.00.645015
max0.5160690.594530368.5711805.073.6250401.765132e+09NaN1.0388.150608388.15060826528.000000388.1506081.00.6866410.6902320.00.699247
\n", "
" ], "text/plain": [ " CER WER TIME PAGES TIME_PER_PAGE timestamp \\\n", "count 64.000000 64.000000 64.000000 64.0 64.000000 6.400000e+01 \n", "mean 0.052482 0.142770 347.605870 5.0 69.423734 1.765126e+09 \n", "std 0.110269 0.107515 7.876539 0.0 1.574470 3.473487e+03 \n", "min 0.011535 0.098902 320.966205 5.0 64.095210 1.765120e+09 \n", "25% 0.011968 0.100441 344.239116 5.0 68.755118 1.765123e+09 \n", "50% 0.012314 0.102033 346.419682 5.0 69.188875 1.765126e+09 \n", "75% 0.040339 0.132047 350.144563 5.0 69.930173 1.765129e+09 \n", "max 0.516069 0.594530 368.571180 5.0 73.625040 1.765132e+09 \n", "\n", " checkpoint_dir_name training_iteration time_this_iter_s \\\n", "count 0.0 64.0 64.000000 \n", "mean NaN 1.0 367.715945 \n", "std NaN 0.0 8.011554 \n", "min NaN 1.0 341.071264 \n", "25% NaN 1.0 364.708660 \n", "50% NaN 1.0 366.103412 \n", "75% NaN 1.0 370.648662 \n", "max NaN 1.0 388.150608 \n", "\n", " time_total_s pid time_since_restore \\\n", "count 64.000000 64.000000 64.000000 \n", "mean 367.715945 16306.750000 367.715945 \n", "std 8.011554 8179.917114 8.011554 \n", "min 341.071264 1104.000000 341.071264 \n", "25% 364.708660 9272.000000 364.708660 \n", "50% 366.103412 18522.000000 366.103412 \n", "75% 370.648662 23167.000000 370.648662 \n", "max 388.150608 26528.000000 388.150608 \n", "\n", " iterations_since_restore config/text_det_thresh \\\n", "count 64.0 64.000000 \n", "mean 1.0 0.419091 \n", "std 0.0 0.167178 \n", "min 1.0 0.016997 \n", "25% 1.0 0.328652 \n", "50% 1.0 0.465068 \n", "75% 1.0 0.530501 \n", "max 1.0 0.686641 \n", "\n", " config/text_det_box_thresh config/text_det_unclip_ratio \\\n", "count 64.000000 64.0 \n", "mean 0.392965 0.0 \n", "std 0.195419 0.0 \n", "min 0.000242 0.0 \n", "25% 0.230515 0.0 \n", "50% 0.448332 0.0 \n", "75% 0.544563 0.0 \n", "max 0.690232 0.0 \n", "\n", " config/text_rec_score_thresh \n", "count 64.000000 \n", "mean 0.470584 \n", "std 0.219216 \n", "min 0.002891 \n", "25% 0.311325 \n", "50% 0.559640 \n", "75% 0.645015 \n", "max 0.699247 " ] }, "execution_count": 7, "metadata": {}, "output_type": "execute_result" } ], "source": [ "#df = pd.read_csv(\"raytune_paddle_subproc_results_20251207_192320.csv\")\n", "df.describe()" ] }, { "cell_type": "code", "execution_count": 9, "id": "50fa5b59", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Correlación con CER:\n", " CER 1.000000\n", "config/text_det_box_thresh 0.226375\n", "config/text_rec_score_thresh -0.160833\n", "config/text_det_thresh -0.522705\n", "config/text_det_unclip_ratio NaN\n", "Name: CER, dtype: float64\n", "Correlación con WER:\n", " WER 1.000000\n", "config/text_det_box_thresh 0.226714\n", "config/text_rec_score_thresh -0.172597\n", "config/text_det_thresh -0.521391\n", "config/text_det_unclip_ratio NaN\n", "Name: WER, dtype: float64\n" ] } ], "source": [ "param_cols = [\n", " \"config/text_det_thresh\",\n", " \"config/text_det_box_thresh\",\n", " \"config/text_det_unclip_ratio\",\n", " \"config/text_rec_score_thresh\",\n", "]\n", "# Correlación de Pearson con CER y WER\n", "corr_cer = df[param_cols + [\"CER\"]].corr()[\"CER\"].sort_values(ascending=False)\n", "corr_wer = df[param_cols + [\"WER\"]].corr()[\"WER\"].sort_values(ascending=False)\n", "\n", "print(\"Correlación con CER:\\n\", corr_cer)\n", "print(\"Correlación con WER:\\n\", corr_wer)" ] }, { "cell_type": "code", "execution_count": 10, "id": "9462b7a2", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "textline_orientation=True:\n", " CER WER\n", "count 53.000000 53.000000\n", "mean 0.037637 0.127337\n", "std 0.098417 0.095844\n", "min 0.011535 0.098902\n", "25% 0.011875 0.100441\n", "50% 0.012199 0.101228\n", "75% 0.012583 0.103280\n", "max 0.516069 0.594530\n", "\n", "textline_orientation=False:\n", " CER WER\n", "count 11.000000 11.000000\n", "mean 0.124009 0.217126\n", "std 0.139431 0.133092\n", "min 0.039052 0.132047\n", "25% 0.044246 0.132462\n", "50% 0.064799 0.164937\n", "75% 0.093873 0.198020\n", "max 0.418109 0.503710\n" ] }, { "data": { "text/plain": [ "" ] }, "execution_count": 10, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# Direct comparison for binary parameter\n", "print(\"textline_orientation=True:\")\n", "print(df[df[\"config/textline_orientation\"] == True][[\"CER\", \"WER\"]].describe())\n", "\n", "print(\"\\ntextline_orientation=False:\")\n", "print(df[df[\"config/textline_orientation\"] == False][[\"CER\", \"WER\"]].describe())\n", "\n", "# Or a simple mean comparison\n", "df.groupby(\"config/textline_orientation\")[[\"CER\", \"WER\"]].mean()\n", "\n", "import seaborn as sns\n", "fig, axes = plt.subplots(1, 2, figsize=(10, 4))\n", "sns.boxplot(data=df, x=\"config/textline_orientation\", y=\"CER\", ax=axes[0])\n", "sns.boxplot(data=df, x=\"config/textline_orientation\", y=\"WER\", ax=axes[1])" ] }, { "cell_type": "markdown", "id": "bc78df46", "metadata": {}, "source": [ "## Interpretation:\n", "\n", "7x better CER with textline_orientation=True. And the variance is much tighter — more reliable results.\n", "For Spanish business documents with mixed layouts (tables, headers, addresses), orientation classification helps PaddleOCR correctly order text lines. Makes sense.\n", "This is thesis-worthy: a single boolean flag accounts for more improvement than all the continuous hyperparameters combined. You could argue that for document OCR pipelines, architectural choices (orientation classification) matter more than threshold tuning." ] }, { "cell_type": "code", "execution_count": 12, "id": "02fc0a87", "metadata": {}, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjcAAAHHCAYAAABDUnkqAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjcsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvTLEjVAAAAAlwSFlzAAAPYQAAD2EBqD+naQAAR3pJREFUeJzt3QmcE/X9//HPshwr13KssIAIAl6ICHIJ9awg3ke1Ra2CVNHiXdQKHiDSigcq/Sveoq1UpSqKWrpqUawHLQpFBRUFUUA5FZZLQJf5P97fXycmIdlNdrObZPJ6Ph5hyWQy853vTGY+870mz/M8zwAAAAKiVroTAAAAkEoENwAAIFAIbgAAQKAQ3AAAgEAhuAEAAIFCcAMAAAKF4AYAAAQKwQ0AAAgUghsAABAoBDdVsHnzZrvgggusuLjY8vLy7Morr3TTV69ebWeccYY1b97cTZ84caJl+zblmscff9xt/5dffmm5mkYtV8vXeiqrffv2duKJJ1qmSHV6ksmj8847z60/G910001uO9etW5fupAAJIbiJc8GI9/r3v/8dmveWW25x8w8fPtyeeOIJO/fcc9303/3ud/bKK6/YqFGj3PRjjz3WUk3rfuGFF6plubG2KRadqP18qVWrljVp0sQOPPBAu/DCC+0///lPldJx3333Vemimu58zCUff/yxu/hlciCI2MrKyuyxxx6zI4880po1a2b16tVzv+uhQ4fa+++/H4hse/LJJ2v8BlP5Ge8ast9++1mmyotKa+PGje2II46wv//971mV/1K7xteYJW6++Wbba6+9dpneqVOn0P9ff/11O+SQQ2zMmDER82j6KaecYldffXW1XpRVOnTqqaemdLnxtimebt262VVXXeX+v2nTJvvkk0/smWeesYcfftgFeXfddVelg5uioiJ3t1ud4uWjgrozzzzTnewzVaakUcHN2LFj3Qk9W0smctH3339vv/jFL6ykpMQOP/xwu+6661yAoyD1b3/7m/35z3+2ZcuW2R577GHZTBfXBQsW1HgptPJt/Pjxu0wvLCy0TDZgwAAbPHiw6bGTX331ld1///120kkn2T/+8Q8bOHBg1uQ/wU0cxx13nPXs2bPczFuzZo117tw55nSVYmSjeNsUT5s2beycc86JmHbbbbfZ2WefbXfffbftvfferhQo2+Tn57tXJsuGNFbFli1brEGDBulORmBdc801LrDR7zT6wqObG02vSbqYbtu2zXbbbTfLdDt37rQdO3ZYQUFB3HkUxESfG6ty3Kcif7Zt22Z169Z1Je3x7LPPPhHpPv3009014U9/+lOlgpu00VPB8ZPHHntMT0n33nvvvbjZ8sYbb7h5ol/+d6NfvvXr13tXXHGFt8cee3h169b1Onbs6N16661eWVlZxPL1fuLEiV6XLl28evXqeUVFRd7AgQNDaYq1jiFDhpS7G1evXu395je/8Vq0aOGW2bVrV+/xxx+vcJuWLl0ad5nt2rXzTjjhhJifbdq0yWvWrJnXpk0bb+fOnRHbdvfdd3udO3d26VB6LrzwQu+7776LWG50Oo444ogazUd/X0Zv/6RJk1zatd5WrVp5F198sUtPOKX1gAMO8BYuXOgdeeSR3m677ea1bt3au+2227xEaL2XXHKJN2XKFG+fffZxaT/44IO9N998M2K+6DTOnDnTy8vL82688caI+f7617+6+e67777QtBUrVnhDhw51+a9t0TY9+uijEd/Tcv3jOp54x7yOp/Bj5K233vJ69erltmWvvfby/vznP8dczqxZs7zhw4d7u+++u9ekSZPQ5zNmzPAOPfRQr379+l7Dhg29448/3luwYEHEMlauXOmdd9557pjTNhUXF3snn3xyxD5MND2yZMkS74wzzvCaNm3q9mGfPn28l19+OaE8ev75590xoOXr77Rp09yxpfUnorqPs+XLl3u1a9f2BgwYkFB6xowZ47bz888/d9tRWFjoNW7c2OX3li1bIuadPHmyd9RRR7l9qPTvv//+Ecde9L4oKSnxevTo4fJK54ZkluEfG4cffrg7Lho1auT17NnTHfN+HkUfm+H7YNu2bd7o0aPdOUTr0TnlmmuucdPj/Sa1X5R32sfx+Psm0XzVPjzrrLPcMd+tW7cK8yeRY/ON/53Tn3rqKe/66693x4bOD9HHUaztjKZzp85F4V544QX3O9Txqbzr0KGDd/PNN3s//vhjRD6kIv8rg5KbOEpLS3dpPKc6SDUS3n///V17FFW7qOjRr5bp3r17qJ2KX7Tn27p1q6u7/Prrr+2iiy6yPffc0959913XLmflypURdZLnn3++a2+i0iM17v3xxx/trbfecu19VJqkdWh67969XfsW6dixY7nFz6oyWLx4sV166aWuuk1VR6ry2bBhg11xxRVxt2n33XevVNDcsGFDO+200+zRRx911RYHHHCAm65t17apTv/yyy+3pUuX2r333mv//e9/7Z133rE6deq4vLjsssvcMq6//nr3vZYtW6Y9H9WuRNUv/fv3d6VRixYtckW27733XijtvvXr17u2Vir2/9WvfmXPPvusXXvtta5NktJTkTfffNOmTp3q8kjVTqqm0/LmzJljXbp0ifmdn//853bxxRe7onBVsx188MEuT5SXSvNvf/vbUIN3VT3qeNbxoH2sImfl18aNG5MqPlZ1htL4//7f/3PVGjqOxP8rOu5U9aflDxkyxCZPnuyOvR49eoSOC5/Sr/SMHj3a3cGK9pO+p7tGlQrqGFC+H3rooe648avCdIe5cOFCt72aplLI1157zVWthFeXJZIe5VG/fv3curR9+t2rmubkk092+1LHdjyvvvpq6G5X++Lbb791x3ui1Ts1cZxpf+v3UF6buli0Dp0/tF3z5s2zRx55xFq0aOH2i09pVT4qr2rXrm0vvfSS268q7bjkkksilqdtO+uss9xvediwYbbvvvsmtQz9vn/zm9+4eXUOUIm5jgmVSKn0WOcPnctXrFgRKonSeUW0LC3/7bffdr9/HbMfffSRm++zzz7bpS2equxVXaffjKrMK6qCVXumWA2wVfISXTLzy1/+0pVyq5r8/2KM+PmT7LE5btw4V1qjZhLbt293/0+G8k/HWfS5UXmvvBwxYoT7q/zR71bnkDvuuMPNk8r8T1qVw6OAiXcnqpci50RKLmJFv+PGjfMaNGjgffbZZxHTR44c6eXn53vLli1z719//XX3/csvv3yX5YaXgGhZFZXW+FR6oWXqrsO3Y8cOr2/fvu5uZ+PGjRVuUywVzau7DK13+vTp7r3ulvXev6vy6c4kerruesJLa2o6H6NLRdasWePuLI455piIEqJ7773Xzac7zei7lb/85S+hadu3b3clCaeffrpXEf94e//990PTvvrqK6+goMA77bTT4qZRdBfdqVMnl3+6+9H+0R22vu87//zz3d3WunXrItZ75plnujvyrVu3JlxyI88880xEaU04vxTuX//6V2ia8lK/pauuumqXbVHpTPidn0oAdTc7bNiwiOWuWrXKpdWfrrtRff+OO+4oN62JpufKK6908+mYDU+LSnnat28fOgZi5ZHuvJW/GzZsCE179dVXd7lrjaWmjrPf/e537rv//e9/vUT4JQwq/Q2n47F58+YR0/zjJ5xKTHVnH2tf6PcfLZFlKH9VUqNSi++//z7ub1y/gVj5/sQTT3i1atWK2MfywAMPuHS98847oWl6r3lVwpKIWCUW/uuiiy7aJV9VahMtXv4kemy+8b+SG+VZrPyMRfPr/LB27Vp3LOocdOyxx8b8bcVaprZNpavhJS+pyP/KoLdUHJMmTXJ3feEv3e1UlkpKDjvsMGvatKmL5v2X7s4U4f/rX/9y8z333HPujjpWg15Nr4wZM2a4rt26A/Dp7k9Rv7p+q5SgOvgRuhoa+3mgemiVaoXnge6YNe8bb7yRsfn4z3/+09Wxq1QjvL5ad1PqURDdm0DbE15vrbsllRB98cUXCa2vb9++Ll98KqFSI3X1wtN2xlO/fn13R6WG3SpVUbp0J6Tvi85fyhs1ENT/w/NQJSO6y9IdeSqpBEP7zKeSGd2BxsoL5Wd4OyL97lS6qGM3PK2ap0+fPqFjRnfDyuNZs2a5u8yqpke/Ge0vlQ6F71PdYarBrUojY1FJ2fz5812JUHjDUR3zibRlq6njTHfX0qhRI0uGX/rnUz6qZMpfnoS3CfFLwFXaqjTpfTiVAsVqx5HIMnRs6NwycuTIXdq+JPIb17lEpQXqvRR+bKkEVKLPR1p/Mu0RVbITfQ3RK1bJaHS+lpc/yR6bQ4YMSaqdjkrb9ZtQiZxKuGfOnGm///3vXQlNuPBlaj8o73Q8qETp008/TXn+J4tqqTh08FTUoDgZn3/+uX344Ydxq3lUhC5Lliyx1q1bu14LqaIW7yryjG5E5lcd6PPqoMAp/ASqPNCJST+a8vIgE/PRzyO/2Dz8YtKhQ4dd8lBVENEnWAVkSnsitL9iNfTTiWPt2rUuWI3nZz/7mavOUICuE6OK7X36roKFhx56yL0qux+S4QdW0XkRKwiJ7qGo/S3+CS+aLviiqjtVjag6VVWYqnbTeDaqGo7Oq0TSo/2p4Cla+G8mVvWgfxzE2n86dioKHGvqOPPzzb/xSFR03mldorzzl6mqM91UzJ492x2v4fT7Dw/6YvVITXQZ+o1LvGraiujY0k1AReeSitIaj6qedNOViHjLjjU92WNzryTTrZsoVb0pyFZVqKrKtA+irx+qAr7hhhtcdVR4cCvRQWwq8j9ZBDc1RPWLuntTBByLLlxBo+5/4d3nlQcKbP7617/GnD+R9j3Zko/xejGF16dXF9WrqwRDdAHQiUklOn7+ie72dUcXS9euXdOWF9F3mH561e4mVkCn9hg+3RGrREp19SrhuvHGG13bEJ181R6uMunJdJXdFn+sFbVx0HAOqVqfjrejjz7aLV/DQLRt29YFZiptUAmivz99sUoUkl1GZWk5apsUb7gKrbeitKZKvGWnYp27JbkMBcx+UHb88ce79kUKdo466ijXtkt0g6SSLAW0GjZF7XFUeqbgXW2+EtlHyeZ/sghuaoh2vkoyKorkNZ9OzN999125pQ7JVK20a9fO3cnpYAqPvv2iQ32eatrW559/3h2g/h2Ftk3F7ipZqOgHF2/70pWPfh6pgZ/uoH26u1Gj6ETv0BLll1iEUyM7BSkVBYG649Ud0YQJE9yJRsX2avAr+q5K0lS1lao0V7a6NBF+I0YFxYmkV/Or9EYv5aEu3HfeeadNmTIlqfVqf2tfR6voN+NPj7X/Yi0vXceZGhsrUFG+JNuouDxq+Kvg+sUXX4wo5UmmiiHRZfjHhm6iwscfS+Zc8sEHH7hAqjqP4VSr7LFZWWrMrKBSpTRqrKy80s2TqiOnTZvmqr99OkYzJf9pc1ND1MtARay64EZTFKyeC6JeFroLUm+J8u7GVOSp7yVC0feqVatc7xuf1nfPPfe4ulpF4Kmk3lk6YSqwUGt5/8BVHuiiqtb70ZSe8O2Jt33pykddVHT3qCAh/Puqn1YR7AknnGCppG0Mr8JYvny5TZ8+3Y455phyx7bRyNAKalSKoQu8xjJRbzS/XZW+q7xRuxu/ZC2cqq2S5ff8SPR4TIaq1XR3qKLxH374IW56VTqlMTyiT54K5HShTJZ+M+qZpv3gU+8tVeWpLUW8thetWrVyAZV6r4QXzautRbx2Ouk4znTToXY86tml80A03QgpKFQvl2T4x2Z42pVujYKc6mXot6D9q9K56H0f/RuPVU2ic4l6XWrA0VjnML+3Xqap7LFZWSod1blEN0w6B8XbRwrA1aszWrryn5KbONR4OFajKHXBC7+jSpQuMroTUTsAv9updp6KhdV9Tw3BVPynoj8FBjq56e5P3Tx1olEXZn2m4kHR91UKoiI9tS1RvWqselhRQ7MHH3zQrXfu3LnuB6B1ql5bXaeTbVQYTgenf1esEhWdwNVQTMGUfhCK+n0KovReJyM1utTJSQ2btZ36jgaJUhddf/vUHfQPf/iDuyvTnbvaXaQrH1Xioa6mCpa0LHVh1N2Tfsy9evWq1GBd5VGduS7s4V3BJVaw5tMJXlVNau/xxz/+MTS/7oTVFVl5pBPNrbfe6u6CtZ26wOlkqEBUwZTyQv9Phi7mOtmpzYtOYkqv9lW8tlXJUGCj40D7Ul3bNSKz9oW6d6txrUoBFbypVEt3gDphant0QlbJobrN6jvJUmnXU0895Uo4tA9U+qeARXemCgzLGwRNx7eCEDX4VHsn5acCCHVX9tuhxVOTx5mCF1UBaft0B67flNrQKG/1e9T5L9m8029awZmqB/Vb1/bq4qVjQY2tU7kMHRsqUdBwDsobdf1W+lUaoGBX+8v/jevGTg1iNZ9u6LRsHVPq2q3GvPo96FjSzZe2W9N1A1WVdpf6LcQrMazKfqzKsVlZOteqm7d+4xpmQtdB5bXON0qDbmBVdRyrOjRd+U9X8CS6gkd3+UymK7jfXW/UqFGuq666e2pgpH79+nkTJkxwXbN96gqrbnf77befm08DWR133HHe3LlzQ/N8+umnbuAqDeCU6CB+GrRN69QyDzzwwJhdfJPtCu7niwaHUpdjdUFW99z//Oc/cb/30EMPuUGplHZ15VRafv/733vffPNNRFdfpUOfRw/iVxP5GG8QP3XJ1fLq1KnjtWzZ0g04F29wtWiJDuIWPmDY3nvv7bopd+/efZeu1tFpVPdedYePznt159SgY0pr+PGgdbRt29Zti7oPH3300W7f+BLtCi4PP/yw63Kq9ccaxC+a8ih8n1Y0eKaWp67A6v6tLvEa9EsDyPnd5dWtXdujfaPu/ZpPXYT/9re/RSwn0fSED5SmruhaZ+/evRMexO+5555zA89p32nQt2QH8auJ48z/jTzyyCPeYYcd5vJM69N3da4I7ybud1lWF+FwsX4nL774ohskVHmmrskaVFBd2KPnK+9ck+gy/Hn1+9dvWOcg7ScNXOfbvHmzd/bZZ7v9GN0dX+cLLdsfcFGD4uncNHbsWK+0tLTCc3pluoKHX3bj5WtF+ZPIsfnG/7qCa6iGRJW3nTfddFPEb1tdtQ855JDQ4JE6h7/yyiu7DAuRivyvjLz/bRCADKG7IA1UphIJAEDyaHMDAAACheAGAAAECsENAAAIFHpLARmGZnAAUDWU3AAAgEAhuAEAAIGSc9VSGsjtm2++cQPXZdOQ2wAA5HqV/aZNm9yAqxUNVphzwY0Cm6o+kAsAAKSHHkejB3yWJ+eCG/9RA8ocDd8NAAAy38aNG13hRCKPDMq54MavilJgQ3ADAEB2SaRJCQ2KAQBAoBDcAACAQCG4AQAAgUJwAwAAAoXgBgAABArBDQAACBSCGwAAECgENwAAIFAIbgAAQKDk3AjF1aVsp2dzln5nazZtsxaNCqz3Xs0svxYP5gQAoKYR3KRAyYKVNvalj21l6bbQtFaFBTbmpM52bJdWqVgFAABIENVSKQhshk+ZFxHYyKrSbW66PgcAADWH4KaKVVEqsfFifOZP0+eaDwAA1AyCmypQG5voEptwCmn0ueYDAAA1g+CmCtR4OJXzAQCAqiO4qQL1ikrlfAAAoOoIbqpA3b3VKypeh29N1+eaDwAA1AyCmyrQODbq7i3RAY7/Xp8z3g0AADWH4KaKNI7N/eccbMWFkVVPeq/piYxzo95Us5d8a9Pnf+3+0rsKAIDKYxC/FFAAM6BzcaVGKGYAQAAAAlhyM2nSJGvfvr0VFBRYnz59bM6cOXHnffzxxy0vLy/ipe+lmwKZvh2b2ynd2ri/iQY2DAAIAEDAgpupU6faiBEjbMyYMTZv3jw76KCDbODAgbZmzZq432ncuLGtXLky9Prqq68s2zAAIAAAAQ1u7rrrLhs2bJgNHTrUOnfubA888IDVr1/fJk+eHPc7Kq0pLi4OvVq2bGnZhgEAAQAIYHCzY8cOmzt3rvXv3/+nBNWq5d7Pnj077vc2b95s7dq1s7Zt29opp5xiCxcutGzDAIAAAAQwuFm3bp2VlZXtUvKi96tWrYr5nX333deV6kyfPt2mTJliO3futH79+tmKFStizr99+3bbuHFjxCsTMAAgAAABrZZKVt++fW3w4MHWrVs3O+KII2zatGm2++6724MPPhhz/vHjx1thYWHopdKeTMAAgAAABDC4KSoqsvz8fFu9enXEdL1XW5pE1KlTx7p3726LFy+O+fmoUaOstLQ09Fq+fLllAgYABAAggMFN3bp1rUePHjZz5szQNFUz6b1KaBKhaq2PPvrIWrWKPVhevXr1XO+q8FeQBgAEAAAZNoifuoEPGTLEevbsab1797aJEyfali1bXO8pURVUmzZtXPWS3HzzzXbIIYdYp06dbMOGDXbHHXe4ruAXXHCB5doAgAAAIAODm0GDBtnatWtt9OjRrhGx2tKUlJSEGhkvW7bM9aDyrV+/3nUd17xNmzZ1JT/vvvuu60aerfwBAAEAQNXleZ7nWQ5Rbyk1LFb7m0yqogIAAKm5fmddbykAAIDyENwAAIBAIbgBAACBQnADAAACheAGAAAECsENAAAIFIIbAAAQKAQ3AAAgUAhuAABAoBDcAACAQCG4AQAAgUJwAwAAAoXgBgAABArBDQAACBSCGwAAECgENwAAIFAIbgAAQKAQ3AAAgEAhuAEAAIFCcAMAAAKF4AYAAAQKwQ0AAAgUghsAABAoBDcAACBQCG4AAECgENwAAIBAIbgBAACBQnADAAACheAGAAAECsENAAAIFIIbAAAQKAQ3AAAgUAhuAABAoBDcAACAQCG4AQAAgUJwAwAAAoXgBgAABArBDQAACBSCGwAAECgENwAAIFAIbgAAQKAQ3AAAgEAhuAEAAIFSO90JQGqU7fRsztLvbM2mbdaiUYH13quZ5dfKI3sBADmH4CYAShastLEvfWwrS7eFprUqLLAxJ3W2Y7u0SmvaAACoaVRLBSCwGT5lXkRgI6tKt7np+hwAgFxCcJPlVVEqsfFifOZP0+eaDwCAXEFwk8XUxia6xCacQhp9rvkAAMgVBDdZTI2HUzkfAABBQHCTxdQrKpXzAQAQBAQ3WUzdvdUrKl6Hb03X55oPAIBcQXCTxTSOjbp7S3SA47/X54x3AwDIJQQ3WU7j2Nx/zsFWXBhZ9aT3ms44NwCAXMMgfgGgAGZA52JGKAYAgOAmOFT11Ldj83QnAwCAtKNaCgAABEpGBDeTJk2y9u3bW0FBgfXp08fmzJmT0Peefvppy8vLs1NPPbXa0wgAALJD2oObqVOn2ogRI2zMmDE2b948O+igg2zgwIG2Zs2acr/35Zdf2tVXX22HHXZYjaUVAABkvrQHN3fddZcNGzbMhg4dap07d7YHHnjA6tevb5MnT477nbKyMvv1r39tY8eOtQ4dOtRoegEAQGZLa3CzY8cOmzt3rvXv3/+nBNWq5d7Pnj077vduvvlma9GihZ1//vkVrmP79u22cePGiBcAAAiutAY369atc6UwLVu2jJiu96tWrYr5nbffftseffRRe/jhhxNax/jx462wsDD0atu2bUrSDgAAMlPaq6WSsWnTJjv33HNdYFNUVJTQd0aNGmWlpaWh1/Lly6s9nQAAIEcH8VOAkp+fb6tXr46YrvfFxcW7zL9kyRLXkPikk04KTdu5c6f7W7t2bVu0aJF17Ngx4jv16tVzLwAAkBvSWnJTt25d69Gjh82cOTMiWNH7vn377jL/fvvtZx999JHNnz8/9Dr55JPtqKOOcv+nygkAAKT98QvqBj5kyBDr2bOn9e7d2yZOnGhbtmxxvadk8ODB1qZNG9d2RuPgdOnSJeL7TZo0cX+jp2ezsp0ej1IAACBbg5tBgwbZ2rVrbfTo0a4Rcbdu3aykpCTUyHjZsmWuB1WuKFmw0sa+9LGtLN0WmtaqsMA93ZuHYAIAULE8z/M8yyHqCq5eU2pc3LhxY8u0wGb4lHkWvUPy/veXp3wDAHLVxiSu37lTJJIFVVEqsYkVafrT9LnmAwAA8RHcZIg5S7+LqIqKppBGn2s+AAAQH8FNhlizaVtK5wMAIFcR3GSIFo0KUjofAAC5iuAmQ/Teq5nrFeU3Ho6m6fpc8wEAgPgIbjJEfq08191bogMc/70+13wAACA+gpsMonFs1N27uDCy6knv6QYOAECWDOKHXQOcAZ2LGaEYAIBKIrjJQKp66tuxebqTAQBAVqJaCgAABArBDQAACBSCGwAAECgENwAAIFAIbgAAQKAQ3AAAgEAhuAEAAIFCcAMAAAKF4AYAAAQKwQ0AAAgUghsAABAoBDcAACBQCG4AAECgENwAAIBAIbgBAACBQnADAAACheAGAAAECsENAAAIFIIbAAAQKAQ3AAAgUAhuAABAoBDcAACAQCG4AQAAgUJwAwAAAoXgBgAABArBDQAACBSCGwAAECgENwAAIFAIbgAAQKAQ3AAAgEAhuAEAAIFCcAMAAAKF4AYAAAQKwQ0AAAgUghsAABAoBDcAACBQCG4AAECgENwAAIBAIbgBAACBQnADAAACheAGAAAECsENAAAIFIIbAAAQKAQ3AAAgUAhuAABAoBDcAACAQCG4AQAAgZIRwc2kSZOsffv2VlBQYH369LE5c+bEnXfatGnWs2dPa9KkiTVo0MC6detmTzzxRI2mFwAAZK60BzdTp061ESNG2JgxY2zevHl20EEH2cCBA23NmjUx52/WrJldf/31Nnv2bPvwww9t6NCh7vXKK6/UeNoBAEDmyfM8z0tnAlRS06tXL7v33nvd+507d1rbtm3tsssus5EjRya0jIMPPthOOOEEGzduXIXzbty40QoLC620tNQaN25c5fQDAIDql8z1O60lNzt27LC5c+da//79f0pQrVruvUpmKqK4bObMmbZo0SI7/PDDY86zfft2lyHhLwAAEFxpDW7WrVtnZWVl1rJly4jper9q1aq431PU1rBhQ6tbt64rsbnnnntswIABMecdP368i/T8l0qFAABAcKW9zU1lNGrUyObPn2/vvfee/fGPf3RtdmbNmhVz3lGjRrlgyH8tX768xtMLAABqTm1Lo6KiIsvPz7fVq1dHTNf74uLiuN9T1VWnTp3c/9Vb6pNPPnElNEceeeQu89arV8+9AABAbkhryY2qlXr06OHazfjUoFjv+/btm/By9B21rQEAAEhryY2oSmnIkCFu7JrevXvbxIkTbcuWLa57twwePNjatGnjSmZEfzVvx44dXUAzY8YMN87N/fffn+YtAQAAmSDtwc2gQYNs7dq1Nnr0aNeIWNVMJSUloUbGy5Ytc9VQPgU+F198sa1YscJ2220322+//WzKlCluOQAAAGkf56amMc4NAADZJ2vGuQEAAEg1ghsAABAoBDcAACBQCG4AAECgENwAAIBASWlws23bNpswYUIqFwkAAFC9wY3GpHn55Zft1VdfdQ+9lB9++MH+9Kc/Wfv27e3WW29NdpEAAADpGcTv7bffthNPPNH1Nc/Ly3MjBT/22GN26qmnWu3ate2mm25yow0DAABkRcnNDTfcYMcff7x9+OGH7rEJeir3aaedZrfccot9/PHH9tvf/taNGgwAAJAVIxQ3b97c3nrrLevcubN9//331rBhQ5s2bZqdcsopli0YoRgAgOxTbSMUr1+/3oqKitz/VUJTv35969KlS9VSCwAAkM4HZ6r6SQ+4FBX6LFq0yD3MMlzXrl1Tl0IAAIDqqpbS07nVkDjWV/zp+uv3ospEVEsBAJB9krl+J1Vys3Tp0qqmDQAAoFolFdy0a9eu+lICAACQAkk1KL799ttdLynfO++8Y9u3bw+937Rpk1188cWpSBcAAED1t7nJz8+3lStXWosWLdx71XnNnz/fOnTo4N6vXr3aWrduTZsbAACQHV3Bo+OgJOIiAACAGsFTwQEAQKAQ3AAAgNwexO+RRx5xj12QH3/80R5//PHQqMVqUAwAAJA1DYrbt2/vBunL5vFwGMQPAIDsU22D+H355ZdVTRsAAEDmtLl5/fXX3RPBFT1FUyR1wAEHuKeGAwAAZEVwM3HiRBs2bFjM4iAVFV100UV21113pTJ9AAAA1RfcfPDBB3bsscfG/fyYY46xuXPnJpcCAACAdAU3GoG4Tp06cT+vXbu2rV27NhXpAgAAqP7gpk2bNrZgwYK4n3/44YfWqlWryqUEAACgpoOb448/3m688Ubbtm3bLp/pgZpjxoyxE088MRXpAgAAqP5xblQtdfDBB7sHaF566aW27777uumffvqpTZo0yT0wc968edayZUvLVIxzAwBA9qm2cW4UtLz77rs2fPhwGzVqVOjBmRrYb+DAgS7AyeTABgAABF/Sj19o166dzZgxw9avX2+LFy92Ac7ee+9tTZs2rZ4UAgAAVGdw41Mw06tXr8p+HQAAoFrwVHAAABAoBDcAACBQCG4AAECgENwAAIBAIbgBAACBQnADAAACheAGAAAECsENAAAIFIIbAAAQKAQ3AAAgUAhuAABAoBDcAACAQCG4AQAAgUJwAwAAAoXgBgAABArBDQAACBSCGwAAECgENwAAIFAIbgAAQKAQ3AAAgEAhuAEAAIFCcAMAAAIlI4KbSZMmWfv27a2goMD69Oljc+bMiTvvww8/bIcddpg1bdrUvfr371/u/AAAILekPbiZOnWqjRgxwsaMGWPz5s2zgw46yAYOHGhr1qyJOf+sWbPsrLPOsjfeeMNmz55tbdu2tWOOOca+/vrrGk87AAC5oGynZ7OXfGvT53/t/up9JsvzPC+tKVRJTa9evezee+9173fu3OkClssuu8xGjhxZ4ffLyspcCY6+P3jw4Arn37hxoxUWFlppaak1btw4JdsAAEBQlSxYaWNf+thWlm4LTWtVWGBjTupsx3ZpVWPpSOb6ndaSmx07dtjcuXNd1VIoQbVqufcqlUnE1q1b7YcffrBmzZrF/Hz79u0uQ8JfAAAgscBm+JR5EYGNrCrd5qbr80yU1uBm3bp1ruSlZcuWEdP1ftWqVQkt49prr7XWrVtHBEjhxo8f7yI9/6VSIQAAUD5VPanEJlb1jj9Nn2diFVXa29xUxa233mpPP/20Pf/8864xciyjRo1yRVj+a/ny5TWeTgAAss2cpd/tUmITTiGNPtd8maZ2OldeVFRk+fn5tnr16ojpel9cXFzudydMmOCCm3/+85/WtWvXuPPVq1fPvQAAQOLWbNqW0vlypuSmbt261qNHD5s5c2ZomhoU633fvn3jfu/222+3cePGWUlJifXs2bOGUgsAQO5o0aggpfPlTMmNqBv4kCFDXJDSu3dvmzhxom3ZssWGDh3qPlcPqDZt2ri2M3LbbbfZ6NGj7cknn3Rj4/htcxo2bOheAACg6nrv1cz1ilLj4VitavLMrLiwwM2XadLe5mbQoEGuikkBS7du3Wz+/PmuRMZvZLxs2TJbufKn1tj333+/62V1xhlnWKtWrUIvLQMAAKRGfq08193bD2TC+e/1uebLNGkf56amMc4NAADBHucm7dVSAAAgcx3bpZUN6FzsekWp8bDa2KgqKhNLbHwENwAAoFwKZPp2bG7ZIu1tbgAAAFKJ4AYAAAQKwQ0AAAgUghsAABAoBDcAACBQCG4AAECgENwAAIBAIbgBAACBQnADAAACheAGAAAECsENAAAIFIIbAAAQKAQ3AAAgUAhuAABAoBDcAACAQCG4AQAAgUJwAwAAAoXgBgAABArBDQAACBSCGwAAECgENwAAIFAIbgAAQKAQ3AAAgEAhuAEAAIFCcAMAAAKF4AYAAAQKwQ0AAAgUghsAABAoBDcAACBQCG4AAECgENwAAIBAIbgBAACBQnADAAACheAGAAAECsENAAAIFIIbAAAQKAQ3AAAgUAhuAABAoBDcAACAQCG4AQAAgUJwAwAAAoXgBgAABErtdCcASIeynZ7NWfqdrdm0zVo0KrDeezWz/Fp57AwACACCG+SckgUrbexLH9vK0m2haa0KC2zMSZ3t2C6t0po2AEDVUS2FnAtshk+ZFxHYyKrSbW66PgcAZDeCG+RUVZRKbLwYn/nT9LnmAwBkL4Ib5Ay1sYkusQmnkEafaz4AQPYiuEHOUOPhVM4HAMhMBDfIGeoVlcr5AACZieAGOUPdvdUrKl6Hb03X55oPAJC9CG6QMzSOjbp7S3SA47/X54x3AwDZjeAGOUXj2Nx/zsFWXBhZ9aT3ms44NwCQ/RjEDzlHAcyAzsWMUAwAAUVwg5ykqqe+HZunOxkAgCBWS02aNMnat29vBQUF1qdPH5szZ07ceRcuXGinn366mz8vL88mTpxYo2kFAACZL63BzdSpU23EiBE2ZswYmzdvnh100EE2cOBAW7NmTcz5t27dah06dLBbb73ViouLazy9AAAg86U1uLnrrrts2LBhNnToUOvcubM98MADVr9+fZs8eXLM+Xv16mV33HGHnXnmmVavXr0aTy8AAMh8aQtuduzYYXPnzrX+/fv/lJhatdz72bNnp2w927dvt40bN0a8AABAcKUtuFm3bp2VlZVZy5YtI6br/apVq1K2nvHjx1thYWHo1bZt25QtGwAAZJ60NyiubqNGjbLS0tLQa/ny5elOEgAACGJX8KKiIsvPz7fVq1dHTNf7VDYWVtsc2ucAAJA70lZyU7duXevRo4fNnDkzNG3nzp3ufd++fdOVLAAAkOXSOoifuoEPGTLEevbsab1793bj1mzZssX1npLBgwdbmzZtXLsZvxHyxx9/HPr/119/bfPnz7eGDRtap06d0rkpAAAgQ6Q1uBk0aJCtXbvWRo8e7RoRd+vWzUpKSkKNjJctW+Z6UPm++eYb6969e+j9hAkT3OuII46wWbNmpWUbAABAZsnzPM9LdyJqkrqCq9eUGhc3btw43ckBAAApvn4HvrcUAADILQQ3AAAgUAhuAABAoBDcAACAQCG4AQAAgUJwAwAAAoXgBgAABEpaB/EDACSmbKdnc5Z+Z2s2bbMWjQqs917NLL9WHtlXCeRl8BHcAECGXxhLFqy0sS99bCtLt4WmtSossDEndbZju7Sq1nUHDXmZGxihGAAy+MKodQ6fMs+ih5L3w6n7zzmYAIe8zAkbGaEYAKonyAgPbGRV6TY3XZ9Hl/DMXvKtTZ//tfur98nSdxRMxfqmP02fV2bZuYa8zC1USwFAFS+MKkXR5wM6F7sqqlSV8Kj6KzqYil63Ptd8fTs2Zz/mYF7Sfig2ghsASOGFsfT7HTGrkfwSnmSqkdSuJ5Xz5bJ052V1BCG0H4qP4AYAUnTBW7Vxm91e8mnCJTwV+XLdloTWq4slUpNH1ZGX1RGExGuLVZkgOogY5wYAUnTB+27z9oRLeBK5039qzrIK59NFUqUAKJ/ySHkVL6TMq6a8TLatViJoP1QxghsASNGFsVmDuikrCVIAtGrj9grnO7PXnox3kwCVlKmkRKL3o/9en6eyW391BSHJVJPmKoIbAEjRhbG4cLeUlQQlWhXWvqh+QvPBXDWNqmuKCyPzX++roxqnuoKQdLcfyga0uQGAJC6M0W0nisPaTugOXCU4qnKIdS+e97/5E6n6SGcbkSDTflKbp5oYiLG6ghCOjYoR3ABAii6MfgmP2lJoileFqg+/KiwVgRIiKf9rort3dQUhHBsVo1oKACpxYTylWxv3NzpQSVXVRzraiCA7GjFzbFSMxy8AQAaPa8JYJtnN7y1lcUryqtLWJ9eOjY1JPH6B4AYAMhyj0Ga36gxCcunY2Ehwk5rMAQAgFXIpCMmE6zcNigEACEgjZvwfGhQDAIBAoeQGAIA0VDvVZFVVWY5VixHcAABQww2Ga7KnU0mO9aoSqqUAAKjBB2NWx8M0U5m+ICC4AQCghh6MWZNP9C6rwXVlGoIbAKgEXRBmL/nWps//2v0N4gUCqX8wZk0+0XtODj89nDY3AJCkXGzDgNQ8GLMmn+i9JoefHk7JDQAkIVfbMCA1D8asySd6t8jhJ8sT3ABAgnK5DQNS82DM6nqYZiy9a3BdmYbgBgASlMttGJCap3PX5BO983P4yfIENwCQoFxuw4DY1MZKT/YuLoys2tH7eE/8rsx3KuvYGlxXJqFBMQAkKJfbMCA+BQgDOhcnNQJwZb5TWcfW4LoyBcENACTZhkGNh2O1qsn73x1xENswIPUPxqzJh2nm59iDO6mWAoAE5XIbBiCbENwAQBJytQ0DkE2olgKAJOViGwYgmxDcAEAl5FobBiCbUC0FAAACheAGAAAECsENAAAIFIIbAAAQKAQ3AAAgUOgtBUdPMU5Ht9bKrDddaU0FP+2rSr+377bssGYN61lx48htqOr2JfL96Hl6tGtqc79an9I8DV9HUYN6bpS7dZu3p32fZdLxEzoeNm6z7zZvt2YN6lpx4W5Zc0zXxPEcvq5/L/nWZn+xzg2ZqJ5qvdo3S/i4rcn9nknHWK5uD8ENrGTBShv70scRTzvWEPMaabU6BySrzHrTldZUiJX26G2QqmxfIvkTax6dp3aGPU+gqnla3rYmuvxELpypSFdx43p2Vu89rX1Rg2oLJhNNSyryf8ePO+2J2V/aV99ttXbN6tu5fdtb3dq1YgYK7y5ZZ99s+N4FVM0a1LGiRgXWolE9+/GHnfb8B1/b1h1lLoAY0u+nZfjbq/Q/M3eFmyde+iV6GxXAjT35ACtqWC/hAHzp2s320Ftf2NYdO0Of3/vGYsvLM/MqOG61nHtfX2yPvbPUNnz/Q9z9nqoAP5vPUanYnkwJhPI8L/zQCL6NGzdaYWGhlZaWWuPGjS3X6cAdPmXeLs/J8Q/F6hpxtTLrTVdaUyFe2hOR6PYlkj+SSDqqkqeJbGtFy6+OC3+i+yDVwWRV8ijZ/B8/42N7+K2lEYGqrivDDtvLRh3fObTukdM+sg1bf7rQV0RBxIWH7WXd92xabtBaWYkE4Aml839//XxLZltTEeBn8zkqFdtT3YFdMtdvgpscpgj70Ntej3sC8R8C+Pa1P09p5F2Z9aYrralQUdoTUdH2JZI/LRvXc/9TFUgq1lnVbY23/Oq48CebrlQFk/HubBNNS6sk8l+BzYP/Whr384sO/7/g5LdT5lmmqUwAXtFxdeMJ+9vFT/43JWlK5DjL5nNUKranJgK7ZIIbGhTnMBUdlneC1UGqzzVfutebrrSmQkVpT0RF25dI/qzauD3hwCaRdVYmHRUtXydU3fklcmHTfJq/OtJV0fLLS2dF308mLYnmv6qiVGJTHn0++oUFlon8XLrpxYV204uJ7f/ylqV8u2F61bY1keMgXDafo6q6PVX5PVQXgpscpjrRVM5XnetNV1pTIZVpires6tzuZJZdmXSEfyfRC3+yF4pk05WKYDLe95NNSyLzq41NRdcNfb5m8w7LVJUJwMvz3ZYfUpKmRI+zbD5HVXV7MjGwI7jJYWrslcr5qnO96UprKqQyTfGWVZ3bncyyK5OO8O9Ux4W/sukqb/lVuZAlm5ZE5lfjYVSfRPZ3Np+jqro9mRjYEdzkMLViV51+vNpfTdfnmi/d601XWlPBT3tVVLR9ieSPeoeot1Gitf2VydOK0lHR8qvjwp9suhJZflUuZMmkJdH8V68o/B/la/MGdVOaHYns72w+R1V1ezIxsCO4yWFqBOZ314w+gP33+jzVjd8qs950pTUV/LQnE1Qku32J5M9NJx9gN50cex5LUZ6Wl45Elp9oIJjshSLRdCW6/KpcyMLTUlEaEs1/dfeuaDZ93qJhai/6qVSZADzWMmTcKV0qFcxW5TjL5nNUVbcnEwO7jAhuJk2aZO3bt7eCggLr06ePzZkzp9z5n3nmGdtvv/3c/AceeKDNmDGjxtIaNGq9rlbsavUeTu+rs9tiZdabrrSmgp/2eBduTX/gnIPdq7Lbl0j+xJsn+nxblTyNt45Elp9MIJjshSKRdKUymCzv+4kcD8nkv8agUXfv8ujzm0/tYjXBP57VQysRlQnAyzuuju/aKqlgtrw0JXOcZfM5qirbk4mBXdq7gk+dOtUGDx5sDzzwgAtsJk6c6IKXRYsWWYsWLXaZ/91337XDDz/cxo8fbyeeeKI9+eSTdtttt9m8efOsS5eKf7iMcxMbIxTXDEYoTmyE4uoa4C58H6j+/8t1W+2pOcsiGrHWxDg31TVCcbrHuTlq3yK78PBOEemf8eE3rudSeAPfisaUqWicm4b1atsFh+5lFx/VqdyB92Itp8ludezQvYvs/S/XR+z3VA5kmSkD2aVKotvDODdhFND06tXL7r33Xvd+586d1rZtW7vsssts5MiRu2TeoEGDbMuWLfbyyy+Hph1yyCHWrVs3FyBVhOAGyA7VMUJxeeupzsdd1KSaGqH4m/Vbbf6KDe7evH3z2OupyuM+Ih7f0bCe63Kzbkvyj++It39q4hEkuaisGn8PWTOI344dO6x+/fr27LPP2qmnnhqaPmTIENuwYYNNnz59l+/sueeeNmLECLvyyitD08aMGWMvvPCCffDBBxWuk+AGAIDsk8z1O63Pllq3bp2VlZVZy5YtI6br/aeffhrzO6tWrYo5v6bHsn37dvcKzxwAABBcGdGguDqpbY4iPf+lKi8AABBcaQ1uioqKLD8/31avXh0xXe+Li4tjfkfTk5l/1KhRrgjLfy1fvjyFWwAAADJNWoObunXrWo8ePWzmzJmhaWpQrPd9+/aN+R1ND59fXnvttbjz16tXz9XNhb8AAEBwpbXNjahxsBoQ9+zZ03r37u26gqs31NChQ93n6ibepk0bV70kV1xxhR1xxBF255132gknnGBPP/20vf/++/bQQw+leUsAAEAmSHtwo67da9eutdGjR7tGwerSXVJSEmo0vGzZMqtV66cCpn79+rmxbW644Qa77rrrbO+993Y9pRIZ4wYAAARf2gfxq2l0BQcAINjX78D3lgIAALmF4AYAAARK2tvc1DS/Fo7B/AAAyB7+dTuR1jQ5F9xs2rTJ/WUwPwAAsvM6rrY35cm5BsUaR+ebb76xRo0aWZ4edVtBlKggSAP/MT4O+cGxwe8kEZw3yAuOi+r5nShcUWDTunXriF7UseRcyY0yZI899kjqOwz+R35wbPA7SRbnDfKC4yL1v5OKSmx8NCgGAACBQnADAAACheCmHHou1ZgxY9xfkB8cG/xOEsF5g7zguEj/7yTnGhQDAIBgo+QGAAAECsENAAAIFIIbAAAQKAQ3AAAgUHI+uJk0aZK1b9/eCgoKrE+fPjZnzpxyM+yZZ56x/fbbz81/4IEH2owZMyxX82PhwoV2+umnu/k12vPEiRMtV/Pi4YcftsMOO8yaNm3qXv3796/wWApqXkybNs169uxpTZo0sQYNGli3bt3siSeesFw+b/iefvpp91s59dRTLRfz4vHHH3fbH/7S93L1uNiwYYNdcskl1qpVK9draJ999gnUNWVSEvlx5JFH7nJs6HXCCSdUbuVeDnv66ae9unXrepMnT/YWLlzoDRs2zGvSpIm3evXqmPO/8847Xn5+vnf77bd7H3/8sXfDDTd4derU8T766CMvF/Njzpw53tVXX+099dRTXnFxsXf33Xd7QZFsXpx99tnepEmTvP/+97/eJ5984p133nleYWGht2LFCi/X8uKNN97wpk2b5n4jixcv9iZOnOh+NyUlJV4QJJsfvqVLl3pt2rTxDjvsMO+UU07xcjEvHnvsMa9x48beypUrQ69Vq1Z5uZgX27dv93r27Okdf/zx3ttvv+2Oj1mzZnnz58/3cjE/vv3224jjYsGCBe68oWOmMnI6uOndu7d3ySWXhN6XlZV5rVu39saPHx9z/l/96lfeCSecEDGtT58+3kUXXeTlYn6Ea9euXaCCm6rkhfz4449eo0aNvD//+c9erueFdO/e3d0MBEFl8kPHQ79+/bxHHnnEGzJkSGCCm2TzQhcqBf1BlGxe3H///V6HDh28HTt2eEHUu4rnDV1PdA7dvHlzpdafs9VSO3bssLlz57rqg/DnTun97NmzY35H08Pnl4EDB8adP+j5EVSpyIutW7faDz/8YM2aNbNczgvdQM2cOdMWLVpkhx9+uGW7yubHzTffbC1atLDzzz/fgqKyebF582Zr166de2jiKaec4qq3czEvXnzxRevbt6+rlmrZsqV16dLFbrnlFisrK7NstyMF59BHH33UzjzzTFe1XRk5G9ysW7fOHUQ6qMLp/apVq2J+R9OTmT/o+RFUqciLa6+91j25NjoYzpW8KC0ttYYNG1rdunVdnfk999xjAwYMsGxXmfx4++233Yla7bKCpDJ5se+++9rkyZNt+vTpNmXKFNu5c6f169fPVqxYYbmWF1988YU9++yz7ntqZ3PjjTfanXfeaX/4wx8s18+hc+bMsQULFtgFF1xQ6TTk3FPBgep26623uoajs2bNClRjyWQ0atTI5s+f7+7SVXIzYsQI69Chg2s0mEs2bdpk5557rgtsioqKLNeppEIvnwKb/fff3x588EEbN26c5RIFdirNe+ihhyw/P9969OhhX3/9td1xxx3usQS57NFHH3Uddnr37l3pZeRscKMTjQ6o1atXR0zX++Li4pjf0fRk5g96fgRVVfJiwoQJLrj55z//aV27drVczQsVQXfq1Mn9X72lPvnkExs/fnzWBzfJ5seSJUvsyy+/tJNOOinioia1a9d21XUdO3a0XD1n1KlTx7p3726LFy+2bFaZvFAPKW2/vudToKeSDVXrqNQzWxVV4djYsmWLuzlUVW5V5Gy1lA4cRcq6qww/6eh9+J1FOE0Pn19ee+21uPMHPT+CqrJ5cfvtt7u7z5KSEtcVOghSdVzoO9u3b7dcyw8NG/HRRx+5Uiz/dfLJJ9tRRx3l/q92J7l8bKjqQvmjC302q0xe/OxnP3NBnR/symeffebyIpsDm6oeGxpuReeKc845x6rEy2HqqlavXj3v8ccfd91WL7zwQtdVze+aeO6553ojR46M6Apeu3Ztb8KECa6775gxYwLXFTyZ/FBXRnV91qtVq1auW7j+//nnn3u5lhe33nqr6/b47LPPRnRn3LRpk5dreXHLLbd4r776qrdkyRI3v34v+t08/PDDXhAkmx/RgtRbKtm8GDt2rPfKK6+4Y2Pu3LnemWee6RUUFLiuwrmWF8uWLXO9gS699FJv0aJF3ssvv+y1aNHC+8Mf/uDl8u/k0EMP9QYNGlTl9ed0cCP33HOPt+eee7oLk7qu/fvf/w59dsQRR7gTUbi//e1v3j777OPmP+CAA7y///3vXq7mh8ZlUHwc/dJ8uZYX6gofKy8UAOdaXlx//fVep06d3EWradOmXt++fd2JLpfPG0ENbpLNiyuvvDI0b8uWLd0YL/PmzfNy9bh499133XAiCgLULfyPf/yjGzYgV/Pj008/dedN3RxVVZ7+qVrZDwAAQObI2TY3AAAgmAhuAABAoBDcAACAQCG4AQAAgUJwAwAAAoXgBgAABArBDQAACBSCGwCV8vjjj1uTJk1yOvfy8vLshRdeqNF16llVWq8e31AV7du3t4kTJ2bc9gGpQHADpNl5553nLiJ66UF6LVu2tAEDBtjkyZMjnjuTiJtuusk9qDLVYl0IBw0a5J6FU930sE0/f/RS/vzyl7+0r776qsbWGf3K9geAAkFHcANkgGOPPdZWrlzp7sr/8Y9/uAcrXnHFFXbiiSfajz/+aJlot912sxYtWtTIuoYNG+by55tvvrHp06fb8uXLq/5gvXJMmzbNrU+vOXPmuGl60rs/TZ9XhgaEz9T9CQQJwQ2QAerVq2fFxcXWpk0bO/jgg+26665zF3EFOqr+8W3YsMEuuOAC23333a1x48b285//3D744AP3meYbO3ase++XMPjfLe97vpdeesl69eplBQUFVlRUZKeddpqbrlIKlZL87ne/Cy03XrXU/fffbx07dnRPBd53333tiSeeiPhc333kkUfcsuvXr2977723vfjiixXmj+ZV/uiJyYcccohdeumlNm/evIh53nzzTevdu7fLS803cuTIUCDxl7/8xRo2bGiff/55aP6LL77YPbV769atu6yvWbNmbn16Kc+kefPmoWn63Ldu3bq42zNr1iy3zdqPekqy0vb222+7Ernx48fbXnvt5YLEgw46yJ599tnQ99avX2+//vWv3br1uZb72GOPRaTxiy++cEGw1qvvz549O+Lz5557zg444AC3TpW83XnnneXmsfLm8MMPd/u/c+fO9tprr1W4X4CMVeWnUwGokvIepHjQQQd5xx13XOh9//79vZNOOsl77733vM8++8y76qqrvObNm3vffvutt3XrVvdeD3T1n0quaRV9T/RE4vz8fG/06NHuCb7z5893T/cWzbPHHnt4N998c2i58thjj3mFhYWhtE2bNs2rU6eON2nSJPeU4zvvvNMt8/XXXw/No1OOlvXkk0+6p8dffvnlXsOGDUPpiEUP2LviiitC7zWvtuWoo44KTVuxYoVXv3597+KLL/Y++eQT7/nnn/eKiooiHlz6y1/+0uvVq5f3ww8/uO1VWt9///0K94//gFg98T5aRdvzxhtvuHm6du3qHga4ePFi95me/Lzffvt5JSUl7gnZyks9PHHWrFnue5dcconXrVs3t7+0/tdee8178cUXI9Kj72s7lNdnnHGGe3irtk20XbVq1XL7TJ9r+bvttpv769P8d999t/t/WVmZ16VLF+/oo492+/7NN9/0unfv7tajvASyDcENkMHBzaBBg7z999/f/f+tt97yGjdu7G3bti1ino4dO3oPPvig+78u5gqIwiXyPT25+9e//nXcNIZfCH3RwU2/fv28YcOGRcyjgEJPfvbpYnnDDTeE3m/evNlN+8c//lFucKNApEGDBi6A0fz77LOPu8j7rrvuOm/ffff1du7cGZqmIEuBhi7c8t1337lAZPjw4e6J1HoCcyIqCm7K2x4/uHnhhRdC82g/aDv0ROhw559/vnfWWWe5/yt4Gzp0aLnpeeSRR0LTFi5c6KYpsJOzzz7bGzBgQMT3rrnmGq9z584x9+krr7zi1a5d2/v6669Dn2sbCG6QraiWAjKYrp9+NZCqkTZv3uyqR1TF4r+WLl1qS5YsibuMRL6nnjdHH310ldL6ySef2M9+9rOIaXqv6eG6du0a+n+DBg1cNdmaNWvKXbaqaJRGbYuqdTp16mTHHHOMbdq0KbTuvn37hvLKX7e2e8WKFe5906ZN7dFHHw1VnanaKhUS2Z6ePXuG/r948WJXFaZG4+H7Q1Vn/v4YPny4Pf30065x+O9//3t79913y12vquHEX2+8faGqp7Kysl2Wpfnbtm1rrVu3Dk1TfgLZqna6EwAgPl101C5DdKHWRUztOKKV1yU7ke+pXUdNUY+wcApIKuoVVlhY6AIa0V8FKdqmqVOnurZEifrXv/5l+fn5rlHwli1brFGjRlYT26OgJ3x/yN///nfXxiqc2sfIcccd59o5zZgxw7V9UeB5ySWX2IQJE2Ku1w/qku1dBwQVJTdAhnr99dfto48+stNPP929V0PjVatWWe3atd0FPvylBsCihrzRd+aJfE+lADNnzoyblljLjbb//vvbO++8EzFN79U4NdUUoMj3338fWrca1P5fTdFP61bwsscee7j3Kv247bbbXMNplZSoUXI6KD8UxCxbtmyX/aHSE58aEw8ZMsSmTJniuuE/9NBDCa8j3r7YZ599QnkXPb96oCno8/373/+u9DYC6UbJDZABtm/f7gIQBRCrV6+2kpIS15tGXcEHDx7s5unfv7+rKjj11FPt9ttvdxcqdY1WCYB666jqQ71iVN2kKhxd1HVxT+R7Y8aMcaUDqq4588wzXS8jlRpce+21bt1arko99JkuzH5QFO6aa66xX/3qV9a9e3e3TgUR6jKtLtRVpWoc5Y8of8aNG+d69ahqyu/5pADgsssuc0HLokWL3DaNGDHCatWq5aqvzj33XLv88stdqYjyRj3DTjrpJDvjjDOsJmmfXH311a73mUpaDj30UCstLXXBh6q0FNCMHj3a9a5SbycdGy+//LILQBJ11VVXue1TPmk8IgV+9957r913330x59f+0nGhdd9xxx22ceNGu/7661O41UANS3ejHyDXqUGxfop6qVHn7rvv7no3TZ48OdQY1rdx40bvsssu81q3bu0a2bZt29Y1BF62bFmoserpp5/uNWnSxC3P7x1T0ffkueeecz106tat63oa/eIXvwh9Nnv2bNfjRz16/NNGdINiue+++7wOHTq4dajR71/+8peIz2M1UNUywnvxxGpQ7OePXk2bNnXTwnthiXoaqTeU0l9cXOxde+21od5Dapx74IEHRjSqVm+uZs2auZ5WVWlQXN72+A2K169fHzGPGj5PnDjRNYJWXmmfDxw40PVSknHjxrmG5OrhpDSqwfkXX3wRNz1avqZpfb5nn33WNSDW8vfcc0/vjjvuKLeRuHpVHXrooS7/tO/Uk4sGxchWefqnpgMqAACA6kKbGwAAECgENwAAIFAIbgAAQKAQ3AAAgEAhuAEAAIFCcAMAAAKF4AYAAAQKwQ0AAAgUghsAABAoBDcAACBQCG4AAECgENwAAAALkv8Ppxpa0o1qbKoAAAAASUVORK5CYII=", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "import matplotlib.pyplot as plt\n", "\n", "plt.scatter(df[\"config/text_det_thresh\"], df[\"CER\"])\n", "plt.xlabel(\"Detection Box Threshold\")\n", "plt.ylabel(\"CER\")\n", "plt.title(\"Effect of Detection pixel threshold on Character Error Rate\")\n", "plt.show()\n", "\n", "plt.scatter(df[\"config/text_det_box_thresh\"], df[\"CER\"])\n", "plt.xlabel(\"Detection Box Threshold\")\n", "plt.ylabel(\"CER\")\n", "plt.title(\"Effect of Detection box threshold on Character Error Rate\")\n", "plt.show()\n", "\n", "plt.scatter(df[\"config/text_det_unclip_ratio\"], df[\"CER\"])\n", "plt.xlabel(\"Detection Box Threshold\")\n", "plt.ylabel(\"CER\")\n", "plt.title(\"Effect of Text detection expansion coefficient on Character Error Rate\")\n", "plt.show()\n", "\n", "plt.scatter(df[\"config/text_rec_score_thresh\"], df[\"WER\"])\n", "plt.xlabel(\"Line Tolerance\")\n", "plt.ylabel(\"WER\")\n", "plt.title(\"Effect of Text recognition threshold on Character Error Rate\")\n", "plt.show()\n", "\n" ] }, { "cell_type": "code", "execution_count": 13, "id": "cc1e3d53", "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "import matplotlib.pyplot as plt\n", "\n", "plt.scatter(df[\"config/text_det_thresh\"], df[\"CER\"])\n", "plt.xlabel(\"Detection Box Threshold\")\n", "plt.ylabel(\"WER\")\n", "plt.title(\"Effect of Detection pixel threshold on Word Error Rate\")\n", "plt.show()\n", "\n", "plt.scatter(df[\"config/text_det_box_thresh\"], df[\"CER\"])\n", "plt.xlabel(\"Detection Box Threshold\")\n", "plt.ylabel(\"WER\")\n", "plt.title(\"Effect of Detection box threshold on Word Error Rate\")\n", "plt.show()\n", "\n", "plt.scatter(df[\"config/text_det_unclip_ratio\"], df[\"CER\"])\n", "plt.xlabel(\"Detection Box Threshold\")\n", "plt.ylabel(\"WER\")\n", "plt.title(\"Effect of Text detection expansion coefficient on Word Error Rate\")\n", "plt.show()\n", "\n", "plt.scatter(df[\"config/text_rec_score_thresh\"], df[\"WER\"])\n", "plt.xlabel(\"Line Tolerance\")\n", "plt.ylabel(\"WER\")\n", "plt.title(\"Effect of Text recognition threshold on Word Error Rate\")\n", "plt.show()\n" ] }, { "cell_type": "code", "execution_count": 17, "id": "1a7e981d", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Best CER: 0.011535\n", "Best WER: 0.098902\n", "\n", "Config:\n", " textline_orientation: True\n", " use_doc_orientation_classify: False\n", " use_doc_unwarping: False\n", " text_det_thresh: 0.4690\n", " text_det_box_thresh: 0.5412\n", " text_det_unclip_ratio: 0.0\n", " text_rec_score_thresh: 0.6350\n" ] } ], "source": [ "best = df.loc[df[\"CER\"].idxmin()]\n", "print(f\"Best CER: {best['CER']:.6f}\")\n", "print(f\"Best WER: {best['WER']:.6f}\")\n", "print(f\"\\nConfig:\")\n", "print(f\" textline_orientation: {best['config/textline_orientation']}\")\n", "print(f\" use_doc_orientation_classify: {best['config/use_doc_orientation_classify']}\")\n", "print(f\" use_doc_unwarping: {best['config/use_doc_unwarping']}\")\n", "print(f\" text_det_thresh: {best['config/text_det_thresh']:.4f}\")\n", "print(f\" text_det_box_thresh: {best['config/text_det_box_thresh']:.4f}\")\n", "print(f\" text_det_unclip_ratio: {best['config/text_det_unclip_ratio']}\")\n", "print(f\" text_rec_score_thresh: {best['config/text_rec_score_thresh']:.4f}\")" ] }, { "cell_type": "markdown", "id": "cfacaf35", "metadata": {}, "source": [ "| Metric | Baseline | Tuned | Improvement |\n", "|--------|----------|-------|-------------|\n", "| CER | 0.01258 | 0.01154 | **-8.3%** |\n", "| WER | 0.10407 | 0.09890 | **-5.0%** |" ] }, { "cell_type": "markdown", "id": "7070a6e6", "metadata": {}, "source": [ "# Graph Interpretatation\n", "\n", "Key insights:\n", "\n", "text_det_thresh (Image 1): Clear failure zone <0.1 (CER 0.4–0.5). Safe range: 0.1–0.7\n", "text_det_box_thresh (Image 2): More scattered, but failures cluster at extremes. Safe range: 0.1–0.5\n", "text_det_unclip_ratio (Image 3): All at 0 (fixed) — confirms that was the right call\n", "text_rec_score_thresh (Image 4): Mid-range values (~0.15–0.2) cause failures. Best at low (<0.1) or high (>0.5)\n", "\n", "Label issues to fix:\n", "\n", "Images 3 & 7: x-axis says \"Detection Box Threshold\" but title says \"expansion coefficient\"\n", "Images 4 & 8: x-axis says \"Line Tolerance\" but title says \"Text recognition threshold\"\n", "\n", "For your thesis, these plots show clear non-linear relationships — you can't just pick defaults. The \"U-shaped\" pattern in text_rec_score_thresh is particularly interesting: both permissive (0) and strict (0.6+) filtering work, but middle values fail.\n", "\n" ] } ], "metadata": { "kernelspec": { "display_name": ".venv (3.11.9)", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.11.9" } }, "nbformat": 4, "nbformat_minor": 5 }