doceker support
This commit is contained in:
68
src/paddle_ocr/Dockerfile.gpu
Normal file
68
src/paddle_ocr/Dockerfile.gpu
Normal file
@@ -0,0 +1,68 @@
|
||||
# Dockerfile.gpu - CUDA-enabled PaddleOCR REST API
|
||||
# Supports: x86_64 with NVIDIA GPU (CUDA 12.x)
|
||||
# For DGX Spark (ARM64 + CUDA): build natively on the device
|
||||
|
||||
FROM nvidia/cuda:12.4.1-cudnn-runtime-ubuntu22.04
|
||||
|
||||
LABEL maintainer="Sergio Jimenez"
|
||||
LABEL description="PaddleOCR Tuning REST API - GPU/CUDA version"
|
||||
|
||||
WORKDIR /app
|
||||
|
||||
# Set environment variables
|
||||
ENV DEBIAN_FRONTEND=noninteractive
|
||||
ENV PYTHONUNBUFFERED=1
|
||||
ENV CUDA_VISIBLE_DEVICES=0
|
||||
|
||||
# Install Python 3.11 and system dependencies
|
||||
RUN apt-get update && apt-get install -y --no-install-recommends \
|
||||
python3.11 \
|
||||
python3.11-venv \
|
||||
python3-pip \
|
||||
libgl1 \
|
||||
libglib2.0-0 \
|
||||
libsm6 \
|
||||
libxext6 \
|
||||
libxrender1 \
|
||||
libgomp1 \
|
||||
&& rm -rf /var/lib/apt/lists/* \
|
||||
&& ln -sf /usr/bin/python3.11 /usr/bin/python
|
||||
|
||||
# Install Python dependencies from requirements file
|
||||
COPY requirements-gpu.txt .
|
||||
RUN pip install --no-cache-dir -r requirements-gpu.txt
|
||||
|
||||
# Copy application code
|
||||
COPY paddle_ocr_tuning_rest.py .
|
||||
COPY dataset_manager.py .
|
||||
|
||||
# Build arguments for models to bake into image
|
||||
ARG DET_MODEL=PP-OCRv5_server_det
|
||||
ARG REC_MODEL=PP-OCRv5_server_rec
|
||||
|
||||
# Set as environment variables (can be overridden at runtime)
|
||||
ENV PADDLE_DET_MODEL=${DET_MODEL}
|
||||
ENV PADDLE_REC_MODEL=${REC_MODEL}
|
||||
|
||||
# Download models during build (not at runtime)
|
||||
RUN python -c "\
|
||||
import os; \
|
||||
from paddleocr import PaddleOCR; \
|
||||
det = os.environ.get('PADDLE_DET_MODEL', 'PP-OCRv5_server_det'); \
|
||||
rec = os.environ.get('PADDLE_REC_MODEL', 'PP-OCRv5_server_rec'); \
|
||||
print(f'Downloading models: det={det}, rec={rec}'); \
|
||||
ocr = PaddleOCR(text_detection_model_name=det, text_recognition_model_name=rec); \
|
||||
print('Models downloaded successfully!')"
|
||||
|
||||
# Volume for dataset and optional additional model cache
|
||||
VOLUME ["/app/dataset", "/root/.paddlex"]
|
||||
|
||||
# Expose API port
|
||||
EXPOSE 8000
|
||||
|
||||
# Health check
|
||||
HEALTHCHECK --interval=30s --timeout=10s --start-period=60s --retries=3 \
|
||||
CMD python -c "import urllib.request; urllib.request.urlopen('http://localhost:8000/health')" || exit 1
|
||||
|
||||
# Run the API server
|
||||
CMD ["uvicorn", "paddle_ocr_tuning_rest:app", "--host", "0.0.0.0", "--port", "8000"]
|
||||
Reference in New Issue
Block a user