updated files

.dockerignore

@@ -12,6 +12,7 @@ venv/
 .env
 
 # Generated / runtime files
+uploadss/
 uploads/
 *.pdf
 *.png

.gitignore

@@ -87,7 +87,7 @@ logs/
 
 
 
-
+uploadss/
 uploads/
 .env
 __pycache__/

notebooks/02_evaluation.ipynb

@@ -0,0 +1,297 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "c13052bd",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# 02_evaluation.ipynb\n",
+    "\n",
+    "## Purpose\n",
+    "\"\"\"\n",
+    "This notebook performs lightweight automated evaluation of the DocVision\n",
+    "pipeline by validating output structure, rule-based constraints, and\n",
+    "measuring latency. It complements the declarative experiments defined\n",
+    "in the experiments/ directory.\n",
+    "\"\"\""
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "id": "3956698b",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import sys\n",
+    "import os\n",
+    "\n",
+    "PROJECT_ROOT = os.path.abspath(os.path.join(os.getcwd(), \"..\"))\n",
+    "if PROJECT_ROOT not in sys.path:\n",
+    "    sys.path.append(PROJECT_ROOT)\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "id": "765c2d64",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import os\n",
+    "import re\n",
+    "import time\n",
+    "import asyncio\n",
+    "from pprint import pprint\n",
+    "\n",
+    "from src.vision import classify_image\n",
+    "from src.textextraction import extract_text_from_image_async\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "id": "075c73df",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Evaluation files loaded: 2\n"
+     ]
+    }
+   ],
+   "source": [
+    "EVAL_FILES = [\n",
+    "    \"C:/Users/bplsy/Documents/DOCVISION_IQ/notebooks/uploadss/driving license .jpg\",\n",
+    "    \"C:/Users/bplsy/Documents/DOCVISION_IQ/notebooks/uploadss/ss.jpg\"\n",
+    "]\n",
+    "\n",
+    "for f in EVAL_FILES:\n",
+    "    assert os.path.exists(f), f\"Missing file: {f}\"\n",
+    "\n",
+    "print(\"Evaluation files loaded:\", len(EVAL_FILES))\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 13,
+   "id": "c94ff0d5",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "async def evaluate_file(path: str):\n",
+    "    start = time.time()\n",
+    "\n",
+    "    # Run OCR explicitly for evaluation\n",
+    "    ocr_text = await extract_text_from_image_async(path)\n",
+    "\n",
+    "    # Run Vision LLM classification\n",
+    "    result = await classify_image(path)\n",
+    "\n",
+    "    latency = round(time.time() - start, 2)\n",
+    "\n",
+    "    checks = {\n",
+    "        \"valid_json\": all(\n",
+    "            k in result for k in [\"document_type\", \"reasoning\", \"extracted_textfields\"]\n",
+    "        ),\n",
+    "        \"latency_sec\": latency,\n",
+    "    }\n",
+    "\n",
+    "    # ✅ CORRECT Aadhaar rule\n",
+    "    if result[\"document_type\"] == \"aadhaar_card\":\n",
+    "        checks[\"aadhaar_rule_passed\"] = bool(\n",
+    "            re.search(r\"\\b\\d{12}\\b\", ocr_text) or \"UIDAI\" in ocr_text.upper()\n",
+    "        )\n",
+    "    else:\n",
+    "        checks[\"aadhaar_rule_passed\"] = \"N/A\"\n",
+    "\n",
+    "    return result, checks\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 14,
+   "id": "1a8d83ed",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\n",
+      "Evaluating: C:/Users/bplsy/Documents/DOCVISION_IQ/notebooks/uploadss/driving license .jpg\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "2026-01-29 13:04:45,587 - INFO - HTTP Request: POST https://api.cloud.llamaindex.ai/api/parsing/upload \"HTTP/1.1 200 OK\"\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Started parsing the file under job_id 8dc83a68-9209-4b25-a474-e4be84fc7aaf\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "2026-01-29 13:04:46,942 - INFO - HTTP Request: GET https://api.cloud.llamaindex.ai/api/parsing/job/8dc83a68-9209-4b25-a474-e4be84fc7aaf \"HTTP/1.1 200 OK\"\n",
+      "2026-01-29 13:04:49,232 - INFO - HTTP Request: GET https://api.cloud.llamaindex.ai/api/parsing/job/8dc83a68-9209-4b25-a474-e4be84fc7aaf \"HTTP/1.1 200 OK\"\n",
+      "2026-01-29 13:04:49,626 - INFO - HTTP Request: GET https://api.cloud.llamaindex.ai/api/parsing/job/8dc83a68-9209-4b25-a474-e4be84fc7aaf/result/text \"HTTP/1.1 200 OK\"\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Error while parsing the file '<bytes/buffer>': Event loop is closed\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "2026-01-29 13:04:52,313 - INFO - HTTP Request: POST https://openrouter.ai/api/v1/chat/completions \"HTTP/1.1 200 OK\"\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\n",
+      "Evaluating: C:/Users/bplsy/Documents/DOCVISION_IQ/notebooks/uploadss/ss.jpg\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "2026-01-29 13:05:04,679 - INFO - HTTP Request: POST https://api.cloud.llamaindex.ai/api/parsing/upload \"HTTP/1.1 200 OK\"\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Started parsing the file under job_id da70766a-4bf1-4341-9beb-3bbe4068374f\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "2026-01-29 13:05:05,961 - INFO - HTTP Request: GET https://api.cloud.llamaindex.ai/api/parsing/job/da70766a-4bf1-4341-9beb-3bbe4068374f \"HTTP/1.1 200 OK\"\n",
+      "2026-01-29 13:05:08,212 - INFO - HTTP Request: GET https://api.cloud.llamaindex.ai/api/parsing/job/da70766a-4bf1-4341-9beb-3bbe4068374f \"HTTP/1.1 200 OK\"\n",
+      "2026-01-29 13:05:08,584 - INFO - HTTP Request: GET https://api.cloud.llamaindex.ai/api/parsing/job/da70766a-4bf1-4341-9beb-3bbe4068374f/result/text \"HTTP/1.1 200 OK\"\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Error while parsing the file '<bytes/buffer>': Event loop is closed\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "2026-01-29 13:05:09,971 - INFO - HTTP Request: POST https://openrouter.ai/api/v1/chat/completions \"HTTP/1.1 200 OK\"\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "C:/Users/bplsy/Documents/DOCVISION_IQ/notebooks/uploadss/driving license .jpg\n",
+      "{'aadhaar_rule_passed': 'N/A', 'latency_sec': 19.71, 'valid_json': True}\n",
+      "C:/Users/bplsy/Documents/DOCVISION_IQ/notebooks/uploadss/ss.jpg\n",
+      "{'aadhaar_rule_passed': True, 'latency_sec': 13.3, 'valid_json': True}\n"
+     ]
+    }
+   ],
+   "source": [
+    "results = await evaluate_all(EVAL_FILES)\n",
+    "\n",
+    "for file, metrics in results:\n",
+    "    print(file)\n",
+    "    pprint(metrics)\n",
+    "\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 15,
+   "id": "9738f8cc",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "All evaluations completed successfully.\n"
+     ]
+    }
+   ],
+   "source": [
+    "print(\"All evaluations completed successfully.\")\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 16,
+   "id": "907d8ab3",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "'\\nThis notebook demonstrates automated evaluation of DocVision outputs,\\nincluding JSON validity, rule-based constraints, and latency measurement.\\nResults support the evaluation protocol described in the README.\\n'"
+      ]
+     },
+     "execution_count": 16,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "## Summary\n",
+    "\"\"\"\n",
+    "This notebook demonstrates automated evaluation of DocVision outputs,\n",
+    "including JSON validity, rule-based constraints, and latency measurement.\n",
+    "Results support the evaluation protocol described in the README.\n",
+    "\"\"\"\n"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": ".venv",
+   "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": "3.13.5"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}