unir/MastersThesis
0
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Imporve data description.

Browse Source
This commit is contained in:
Sergio Jimenez Jimenez
2025-12-08 11:34:23 +01:00
parent 0fd1323cf2
commit 0d3e0760a1
1 changed files with 8 additions and 35 deletions
Download Patch File Download Diff File Expand all files Collapse all files

43
src/paddle_ocr_fine_tune_unir_raytune.ipynb
View File

@@ -2322,7 +2322,7 @@
"max 0.699247 " "max 0.699247 "
] ]
}, },
"execution_count": 20, "execution_count": 32,
"metadata": {}, "metadata": {},
"output_type": "execute_result" "output_type": "execute_result"
} }
@@ -2334,7 +2334,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 28, "execution_count": 33,
"id": "50fa5b59", "id": "50fa5b59",
"metadata": {}, "metadata": {},
"outputs": [ "outputs": [
@@ -2383,44 +2383,17 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 29, "execution_count": 34,
"id": "9462b7a2", "id": "9462b7a2",
"metadata": {}, "metadata": {},
"outputs": [ "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": { "data": {
"text/plain": [ "text/plain": [
"<Axes: xlabel='config/textline_orientation', ylabel='WER'>" "<Axes: xlabel='config/textline_orientation', ylabel='WER'>"
] ]
}, },
"execution_count": 29, "execution_count": 34,
"metadata": {}, "metadata": {},
"output_type": "execute_result" "output_type": "execute_result"
}, },
@@ -2437,11 +2410,11 @@
], ],
"source": [ "source": [
"# Direct comparison for binary parameter\n", "# Direct comparison for binary parameter\n",
"print(\"textline_orientation=True:\")\n", "#print(\"textline_orientation=True:\")\n",
"print(df[df[\"config/textline_orientation\"] == True][[\"CER\", \"WER\"]].describe())\n", "#print(df[df[\"config/textline_orientation\"] == True][[\"CER\", \"WER\"]].describe())\n",
"\n", "\n",
"print(\"\\ntextline_orientation=False:\")\n", "#print(\"\\ntextline_orientation=False:\")\n",
"print(df[df[\"config/textline_orientation\"] == False][[\"CER\", \"WER\"]].describe())\n", "#print(df[df[\"config/textline_orientation\"] == False][[\"CER\", \"WER\"]].describe())\n",
"\n", "\n",
"# Or a simple mean comparison\n", "# Or a simple mean comparison\n",
"df.groupby(\"config/textline_orientation\")[[\"CER\", \"WER\"]].mean()\n", "df.groupby(\"config/textline_orientation\")[[\"CER\", \"WER\"]].mean()\n",