{ "cells": [ { "cell_type": "code", "execution_count": 1, "id": "f65bb771", "metadata": {}, "outputs": [], "source": [ "# Run this as the VERY FIRST thing — before any imports\n", "!pip install -q langchain langchain-community langchain-core langchain-text-splitters langchain-chroma langchain-huggingface langchain-groq chromadb rank_bm25 sentence-transformers pypdf groq pydantic pandas matplotlib seaborn tenacity python-dotenv" ] }, { "cell_type": "code", "execution_count": 2, "id": "1919ed83", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "\u001b[31mERROR: Could not open requirements file: [Errno 2] No such file or directory: 'requirements.txt'\u001b[0m\u001b[31m\n", "\u001b[0mMounting Google Drive...\n", "Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount(\"/content/drive\", force_remount=True).\n", "Module cache flushed (0 src.* modules cleared).\n", "Importing modules...\n", "Working directory: /content/drive/MyDrive/financial-intelligence-engine\n", "All modules imported successfully.\n", "\n", "Skipping PDF ingestion — Smart Load active in Cell 2.\n", "To rebuild indexes from scratch: uncomment the ingestion block in Cell 2.\n" ] } ], "source": [ "# ============================================================\n", "# FINANCIAL INTELLIGENCE ENGINE\n", "# Cell 1 — Environment Setup & Module Import\n", "# ============================================================\n", "\n", "# Install all pinned dependencies from requirements.txt\n", "# This single command replaces multiple scattered !pip install lines.\n", "!pip install -q -r requirements.txt\n", "\n", "import sys\n", "import os\n", "from google.colab import drive\n", "\n", "# ── Mount Google Drive ────────────────────────────────────────\n", "print('Mounting Google Drive...')\n", "drive.mount('/content/drive')\n", "\n", "PROJECT_PATH = '/content/drive/MyDrive/financial-intelligence-engine'\n", "os.chdir(PROJECT_PATH)\n", "\n", "# Force Python to resolve src/ imports from the project root.\n", "if PROJECT_PATH not in sys.path:\n", " sys.path.insert(0, PROJECT_PATH)\n", "\n", "# ── Flush Jupyter Module Cache ────────────────────────────────\n", "# Required in Colab: stale cached module bytecode from previous runs\n", "# causes import errors when source files change between cells.\n", "modules_to_delete = [mod for mod in sys.modules if mod.startswith('src')]\n", "for mod in modules_to_delete:\n", " del sys.modules[mod]\n", "print(f'Module cache flushed ({len(modules_to_delete)} src.* modules cleared).')\n", "\n", "# ── Import & Environment Setup ────────────────────────────────\n", "print('Importing modules...')\n", "from src.config import logger, setup_environment\n", "\n", "# setup_environment() creates all artifact directories.\n", "# Called once here so no other module needs to run side effects on import.\n", "setup_environment()\n", "\n", "from src.data_ingestion import load_and_chunk_pdfs\n", "\n", "print(f'Working directory: {os.getcwd()}')\n", "print('All modules imported successfully.')\n", "print('\\nSkipping PDF ingestion — Smart Load active in Cell 2.')\n", "print('To rebuild indexes from scratch: uncomment the ingestion block in Cell 2.')" ] }, { "cell_type": "code", "execution_count": 5, "id": "3a4dcee9", "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "2026-04-03 05:37:03 - [INFO] - financial_rag - Loading embedding model: BAAI/bge-small-en-v1.5\n", "INFO:financial_rag:Loading embedding model: BAAI/bge-small-en-v1.5\n", "/usr/local/lib/python3.12/dist-packages/huggingface_hub/utils/_auth.py:104: UserWarning: \n", "Error while fetching `HF_TOKEN` secret value from your vault: 'Requesting secret HF_TOKEN timed out. Secrets can only be fetched when running from the Colab UI.'.\n", "You are not authenticated with the Hugging Face Hub in this notebook.\n", "If the error persists, please let us know by opening an issue on GitHub (https://github.com/huggingface/huggingface_hub/issues/new).\n", " warnings.warn(\n", "Warning: You are sending unauthenticated requests to the HF Hub. Please set a HF_TOKEN to enable higher rate limits and faster downloads.\n", "WARNING:huggingface_hub.utils._http:Warning: You are sending unauthenticated requests to the HF Hub. Please set a HF_TOKEN to enable higher rate limits and faster downloads.\n" ] }, { "data": { "application/vnd.jupyter.widget-view+json": { "model_id": "bf15cc96150745f599dd2eddb73e08a2", "version_major": 2, "version_minor": 0 }, "text/plain": [ "Loading weights: 0%| | 0/199 [00:00 {source}')\\n\\n \"" ] }, "execution_count": 7, "metadata": {}, "output_type": "execute_result" } ], "source": [ "\"\"\"\n", "\n", "# ============================================================\n", "# Cell 3 — Generation Agent (CoT + Self-Correction)\n", "# ============================================================\n", "\n", "import os\n", "from dotenv import load_dotenv\n", "from src.generation_agent import FinancialGenerationAgent\n", "\n", "# ── Secure Credential Loading ─────────────────────────────────\n", "print('Loading credentials from Drive .env file...')\n", "load_dotenv(override=True) # override=True ensures Drive .env takes precedence\n", "GROQ_API_KEY = os.getenv('GROQ_API_KEY')\n", "\n", "if not GROQ_API_KEY:\n", " raise ValueError(\n", " 'GROQ_API_KEY not found. '\n", " 'Ensure your .env file is in the project root on Google Drive '\n", " 'and contains: GROQ_API_KEY=your_key_here'\n", " )\n", "print('API key loaded successfully.')\n", "\n", "# ── Initialize Agent ──────────────────────────────────────────\n", "print('\\nPhase 3: Initializing Generation & Self-Correction Agent...')\n", "agent = FinancialGenerationAgent(retriever=ensemble_retriever, api_key=GROQ_API_KEY)\n", "\n", "# ── Run Primary Query ─────────────────────────────────────────\n", "complex_query = (\n", " 'Compare the key areas where Google and Meta are investing their '\n", " 'Research and Development (R&D) budgets for 2025.'\n", ")\n", "\n", "print(f'\\nQuery: {complex_query}')\n", "final_answer, used_docs = agent.generate_answer(complex_query)\n", "\n", "print('\\n' + '='*60)\n", "print(' FINANCIAL INTELLIGENCE ENGINE OUTPUT')\n", "print('='*60)\n", "print(final_answer)\n", "\n", "print('\\n' + '-'*60)\n", "print(' SOURCES CITED')\n", "print('-'*60)\n", "sources = set(d.metadata.get('source_file') for d in used_docs)\n", "for source in sources:\n", " print(f' -> {source}')\n", "\n", " \"\"\"" ] }, { "cell_type": "code", "execution_count": null, "id": "868da12c", "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "code", "execution_count": 9, "id": "651ad1bf", "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "2026-04-03 05:38:39 - [INFO] - financial_rag - Running LLM-as-a-Judge evaluation (Judge model: qwen/qwen3-32b)...\n", "INFO:financial_rag:Running LLM-as-a-Judge evaluation (Judge model: qwen/qwen3-32b)...\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Phase 4a: Single-Query Evaluation (quick check)...\n" ] }, { "name": "stderr", "output_type": "stream", "text": [ "2026-04-03 05:38:40 - [INFO] - financial_rag - Evaluation scores: {'faithfulness': 1.0, 'relevance': 1.0}\n", "INFO:financial_rag:Evaluation scores: {'faithfulness': 1.0, 'relevance': 1.0}\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\n", "=======================================================\n", " SINGLE-QUERY EVALUATION RESULTS\n", "=======================================================\n", " Metric Score (0.0 - 1.0) Status\n", "Faithfulness (Hallucination Check) 1.0 PASS\n", " Context Relevance 1.0 PASS\n", "\n", "Note: Single-query scores are directional only.\n", "Run Cell 5 (Batch Evaluation) for statistically valid aggregate scores.\n" ] } ], "source": [ "# ============================================================\n", "# Cell 4 — Single-Query Evaluation (Quick Check)\n", "# ============================================================\n", "\n", "import pandas as pd\n", "from src.evaluation import RAGEvaluator\n", "\n", "print('Phase 4a: Single-Query Evaluation (quick check)...')\n", "evaluator = RAGEvaluator(api_key=GROQ_API_KEY)\n", "\n", "scores = evaluator.evaluate(\n", " question=complex_query,\n", " answer=final_answer,\n", " context_docs=used_docs,\n", " # Optional: provide a ground_truth string to enable the Correctness metric.\n", " # ground_truth='Google and Meta both invest heavily in AI; ...',\n", ")\n", "\n", "print('\\n' + '='*55)\n", "print(' SINGLE-QUERY EVALUATION RESULTS')\n", "print('='*55)\n", "\n", "metrics_data = [\n", " {\n", " 'Metric': 'Faithfulness (Hallucination Check)',\n", " 'Score (0.0 - 1.0)': scores.get('faithfulness', 'Error'),\n", " 'Status': (\n", " 'PASS' if isinstance(scores.get('faithfulness'), float)\n", " and scores['faithfulness'] >= 0.8 else 'FAIL'\n", " )\n", " },\n", " {\n", " 'Metric': 'Context Relevance',\n", " 'Score (0.0 - 1.0)': scores.get('relevance', 'Error'),\n", " 'Status': (\n", " 'PASS' if isinstance(scores.get('relevance'), float)\n", " and scores['relevance'] >= 0.8 else 'FAIL'\n", " )\n", " },\n", "]\n", "\n", "if 'correctness' in scores:\n", " metrics_data.append({\n", " 'Metric': 'Answer Correctness (vs Ground Truth)',\n", " 'Score (0.0 - 1.0)': scores.get('correctness', 'Error'),\n", " 'Status': (\n", " 'PASS' if isinstance(scores.get('correctness'), float)\n", " and scores['correctness'] >= 0.8 else 'FAIL'\n", " )\n", " })\n", "\n", "metrics_df = pd.DataFrame(metrics_data)\n", "print(metrics_df.to_string(index=False))\n", "print('\\nNote: Single-query scores are directional only.')\n", "print('Run Cell 5 (Batch Evaluation) for statistically valid aggregate scores.')" ] }, { "cell_type": "code", "execution_count": 10, "id": "73e7972c", "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "2026-04-03 05:38:45 - [INFO] - financial_rag - Starting batch evaluation over 15 questions (Judge: qwen/qwen3-32b)...\n", "INFO:financial_rag:Starting batch evaluation over 15 questions (Judge: qwen/qwen3-32b)...\n", "2026-04-03 05:38:45 - [INFO] - financial_rag - Evaluating question 1/15: 'What were Google's total Research and Development expenses i'\n", "INFO:financial_rag: Evaluating question 1/15: 'What were Google's total Research and Development expenses i'\n", "2026-04-03 05:38:45 - [INFO] - financial_rag - Retrieving documents for query: 'What were Google's total Research and Development expenses in fiscal year 2025?'\n", "INFO:financial_rag:Retrieving documents for query: 'What were Google's total Research and Development expenses in fiscal year 2025?'\n", "2026-04-03 05:38:45 - [INFO] - financial_rag - Step 1: Executing Chain-of-Thought Analysis (llama-3.3-70b-versatile)...\n", "INFO:financial_rag:Step 1: Executing Chain-of-Thought Analysis (llama-3.3-70b-versatile)...\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Phase 4b: Running Batch Evaluation over 15 questions...\n", "Judge model : qwen/qwen3-32b\n", "Generator : llama-3.3-70b-versatile\n", "This will take 5-8 minutes. Each question runs full generation + evaluation.\n", "\n" ] }, { "name": "stderr", "output_type": "stream", "text": [ "2026-04-03 05:38:46 - [INFO] - financial_rag - Step 2: Running Strict Compliance Audit (llama-3.3-70b-versatile)...\n", "INFO:financial_rag:Step 2: Running Strict Compliance Audit (llama-3.3-70b-versatile)...\n", "2026-04-03 05:38:46 - [INFO] - financial_rag - Running LLM-as-a-Judge evaluation (Judge model: qwen/qwen3-32b)...\n", "INFO:financial_rag:Running LLM-as-a-Judge evaluation (Judge model: qwen/qwen3-32b)...\n", "2026-04-03 05:38:48 - [ERROR] - financial_rag - Failed to parse LLM evaluation response: Invalid json output: \n", "Okay, let's start by looking at the user's question: What were Google's total Research and Development expenses in fiscal year 2025?\n", "\n", "The context provided has several sections. The first part mentions that in the Research and Development section, there's a table showing expenses for 2024 and 2025 as $49,326 million and $61,087 million respectively. Then later, there's another mention of Research and Development expenses increasing by $13.50 billion in 2025 compared to 2024. Wait, that's conflicting. The first table says an increase of $11.8 billion (from $49,326 to $61,087), but later it says $13.50 billion. Hmm, that's a discrepancy. But the user's answer is based on the first table's $61,087 million figure.\n", "\n", "The answer given is $61,087 million, which is $61.087 billion. The Ground Truth states the same amount, $61.087 billion, with 15% of revenues and an increase of $11.8 billion. So the answer matches the Ground Truth in the amount. \n", "\n", "Now, checking faithfulness: The answer cites the source as the Google 10-K and uses the figure from the first table. The context does have that table. However, there's another part in the context that says \"Research and development expenses in 2025 increased $13.50 billion, or 31%, compared to 2024.\" This is conflicting. The first table shows an increase of $11.8 billion (from $49,326 to $61,087 is $11,761 million, which rounds to $11.8 billion). The later statement says $13.50 billion. Which one is correct? The answer uses the first figure. The Ground Truth also mentions the $11.8 billion increase. So the answer is correct in using the first figure. However, the context has conflicting data. But since the answer is based on the table provided in the context, it's still faithful because it's using the data from the context. The presence of conflicting data in the context might be an issue, but the answer is using the correct part from the context.\n", "\n", "Relevance: The answer directly addresses the question by providing the 2025 R&D expenses. It's completely relevant.\n", "\n", "Correctness: The Ground Truth states $61.087 billion, which matches the answer. The answer is correct. The Ground Truth also mentions the 15% of revenues and the $11.8 billion increase, which the answer doesn't include, but the question only asked for the total expenses. So the answer is correct as per the question's requirements.\n", "\n", "So faithfulness is 1.0 because the answer is derived from the context. Relevance is 1.0. Correctness is 1.0.\n", "\n", "\n", "{\n", " \"faithfulness\": 1.0,\n", " \"relevance\": 1.0,\n", " \"correctness\": 1.0\n", "}\n", "For troubleshooting, visit: https://docs.langchain.com/oss/python/langchain/errors/OUTPUT_PARSING_FAILURE \n", "ERROR:financial_rag:Failed to parse LLM evaluation response: Invalid json output: \n", "Okay, let's start by looking at the user's question: What were Google's total Research and Development expenses in fiscal year 2025?\n", "\n", "The context provided has several sections. The first part mentions that in the Research and Development section, there's a table showing expenses for 2024 and 2025 as $49,326 million and $61,087 million respectively. Then later, there's another mention of Research and Development expenses increasing by $13.50 billion in 2025 compared to 2024. Wait, that's conflicting. The first table says an increase of $11.8 billion (from $49,326 to $61,087), but later it says $13.50 billion. Hmm, that's a discrepancy. But the user's answer is based on the first table's $61,087 million figure.\n", "\n", "The answer given is $61,087 million, which is $61.087 billion. The Ground Truth states the same amount, $61.087 billion, with 15% of revenues and an increase of $11.8 billion. So the answer matches the Ground Truth in the amount. \n", "\n", "Now, checking faithfulness: The answer cites the source as the Google 10-K and uses the figure from the first table. The context does have that table. However, there's another part in the context that says \"Research and development expenses in 2025 increased $13.50 billion, or 31%, compared to 2024.\" This is conflicting. The first table shows an increase of $11.8 billion (from $49,326 to $61,087 is $11,761 million, which rounds to $11.8 billion). The later statement says $13.50 billion. Which one is correct? The answer uses the first figure. The Ground Truth also mentions the $11.8 billion increase. So the answer is correct in using the first figure. However, the context has conflicting data. But since the answer is based on the table provided in the context, it's still faithful because it's using the data from the context. The presence of conflicting data in the context might be an issue, but the answer is using the correct part from the context.\n", "\n", "Relevance: The answer directly addresses the question by providing the 2025 R&D expenses. It's completely relevant.\n", "\n", "Correctness: The Ground Truth states $61.087 billion, which matches the answer. The answer is correct. The Ground Truth also mentions the 15% of revenues and the $11.8 billion increase, which the answer doesn't include, but the question only asked for the total expenses. So the answer is correct as per the question's requirements.\n", "\n", "So faithfulness is 1.0 because the answer is derived from the context. Relevance is 1.0. Correctness is 1.0.\n", "\n", "\n", "{\n", " \"faithfulness\": 1.0,\n", " \"relevance\": 1.0,\n", " \"correctness\": 1.0\n", "}\n", "For troubleshooting, visit: https://docs.langchain.com/oss/python/langchain/errors/OUTPUT_PARSING_FAILURE \n", "2026-04-03 05:38:48 - [INFO] - financial_rag - Evaluating question 2/15: 'What were Google's total revenues and net income in fiscal y'\n", "INFO:financial_rag: Evaluating question 2/15: 'What were Google's total revenues and net income in fiscal y'\n", "2026-04-03 05:38:48 - [INFO] - financial_rag - Retrieving documents for query: 'What were Google's total revenues and net income in fiscal year 2025?'\n", "INFO:financial_rag:Retrieving documents for query: 'What were Google's total revenues and net income in fiscal year 2025?'\n", "2026-04-03 05:38:48 - [INFO] - financial_rag - Step 1: Executing Chain-of-Thought Analysis (llama-3.3-70b-versatile)...\n", "INFO:financial_rag:Step 1: Executing Chain-of-Thought Analysis (llama-3.3-70b-versatile)...\n", "2026-04-03 05:38:49 - [INFO] - financial_rag - Step 2: Running Strict Compliance Audit (llama-3.3-70b-versatile)...\n", "INFO:financial_rag:Step 2: Running Strict Compliance Audit (llama-3.3-70b-versatile)...\n", "2026-04-03 05:38:49 - [INFO] - financial_rag - Running LLM-as-a-Judge evaluation (Judge model: qwen/qwen3-32b)...\n", "INFO:financial_rag:Running LLM-as-a-Judge evaluation (Judge model: qwen/qwen3-32b)...\n", "2026-04-03 05:39:19 - [INFO] - financial_rag - Evaluation scores: {'faithfulness': 1.0, 'relevance': 1.0, 'correctness': 1.0}\n", "INFO:financial_rag:Evaluation scores: {'faithfulness': 1.0, 'relevance': 1.0, 'correctness': 1.0}\n", "2026-04-03 05:39:19 - [INFO] - financial_rag - Evaluating question 3/15: 'How much did Google spend on capital expenditures in fiscal '\n", "INFO:financial_rag: Evaluating question 3/15: 'How much did Google spend on capital expenditures in fiscal '\n", "2026-04-03 05:39:19 - [INFO] - financial_rag - Retrieving documents for query: 'How much did Google spend on capital expenditures in fiscal year 2025?'\n", "INFO:financial_rag:Retrieving documents for query: 'How much did Google spend on capital expenditures in fiscal year 2025?'\n", "2026-04-03 05:39:20 - [INFO] - financial_rag - Step 1: Executing Chain-of-Thought Analysis (llama-3.3-70b-versatile)...\n", "INFO:financial_rag:Step 1: Executing Chain-of-Thought Analysis (llama-3.3-70b-versatile)...\n", "2026-04-03 05:39:20 - [INFO] - financial_rag - Step 2: Running Strict Compliance Audit (llama-3.3-70b-versatile)...\n", "INFO:financial_rag:Step 2: Running Strict Compliance Audit (llama-3.3-70b-versatile)...\n", "2026-04-03 05:39:21 - [INFO] - financial_rag - Running LLM-as-a-Judge evaluation (Judge model: qwen/qwen3-32b)...\n", "INFO:financial_rag:Running LLM-as-a-Judge evaluation (Judge model: qwen/qwen3-32b)...\n", "2026-04-03 05:39:46 - [ERROR] - financial_rag - Failed to parse LLM evaluation response: Invalid json output: \n", "Okay, let's start by looking at the user's question: they want to know how much Google spent on capital expenditures in fiscal year 2025. The context provided includes a lot of financial data from Google's filings. \n", "\n", "First, I need to check if the answer is present in the context. The user's answer states that the capital expenditures for 2025 aren't explicitly mentioned, but the Ground Truth says it's $91.4 billion. Let me scan through the context again.\n", "\n", "Looking through the sections, under \"Cash Used in Investing Activities\" for 2025, there's a line: \"$69.69 billion of purchases of property and equipment...\" Wait, property and equipment purchases are part of capital expenditures. But the Ground Truth mentions $91.4 billion. However, the context also says \"We anticipate making capital expenditures of approximately $115 billion to $135 billion in 2026...\" which is a future year. \n", "\n", "Wait, the user's answer says the 2025 capital expenditures aren't stated, but the Ground Truth claims it's $91.4B. Let me check again. The context under \"Cash Used in Investing Activities during 2025\" lists $69.69B for property and equipment. But capital expenditures can include more than just property and equipment. However, the Ground Truth says $91.4B, which isn't directly stated in the context. The context only mentions $69.69B for property and equipment. \n", "\n", "So the answer correctly points out that the context doesn't explicitly state the total capital expenditures for 2025. The Ground Truth is providing a figure that isn't in the context. Therefore, the answer is faithful because it doesn't make up a number not present. Relevance is high because it directly addresses the question. Correctness is low because the answer doesn't match the Ground Truth. The Ground Truth's $91.4B isn't in the context, so the answer is correct in stating it's not provided. Therefore, the correctness score should be 0.0 because the answer contradicts the Ground Truth by not providing the correct figure when it's actually present in the context? Wait, no. Wait, the context does mention $69.69B for property and equipment under 2025. But the Ground Truth says $91.4B. Is there another line item in the context that adds up to $91.4B? Let me check again.\n", "\n", "In the context under \"Cash Used in Investing Activities during 2025\": $69.69B (property and equipment), $18.33B (non-marketable equity investments), and $10.05B (marketable securities). Adding those gives $98.07B. But the Ground Truth says $91.4B. Hmm, maybe the Ground Truth is incorrect, or perhaps the answer is missing something. Alternatively, maybe the $69.69B is part of the total capital expenditures. But the answer says it's not stated, but the context does have a figure. Wait, the answer says the context doesn't provide a specific amount for capital expenditures, but the context does mention $69.69B for property and equipment. However, capital expenditures can include other items. The answer is correct in stating that the total isn't explicitly given, but the Ground Truth is providing a figure that isn't in the context. Therefore, the answer is faithful (since it doesn't hallucinate a number), relevant (addresses the question), but incorrect compared to the Ground Truth. However, the Ground Truth might be wrong if the context doesn't actually state $91.4B. Wait, the user's Ground Truth says $91.4B, but the context only mentions $69.69B for property and equipment. Unless there's another line item. Let me check again.\n", "\n", "Looking at the context: \"Cash used in investing activities during 2025 mostly consisted of $69.69 billion of purchases of property and equipment...\" So that's part of capital expenditures. But the total cash used in investing activities is $69.69B + $18.33B + $10.05B = $98.07B. However, capital expenditures typically refer to purchases of property and equipment. The other items (non-marketable equity investments and marketable securities) are not usually considered capital expenditures. Therefore, the actual capital expenditures for 2025 would be $69.69B. But the Ground Truth says $91.4B, which is not present in the context. Therefore, the answer is correct in stating that the context doesn't provide the total capital expenditures for 2025, and the Ground Truth is incorrect. However, the user's Ground Truth is given as a reference, so the answer's correctness is 0.0 because it doesn't match the Ground Truth. But wait, the answer is correct based on the context, but the Ground Truth is wrong. However, the user's instruction says to score correctness based on agreement with the Ground Truth. So even if the Ground Truth is incorrect, the answer's correctness is 0.0 because it doesn't match. Therefore, the answer is not correct compared to the Ground Truth. So the scores would be: faithfulness 1.0 (since it doesn't make up a number), relevance 1.0 (directly answers the question), correctness 0.0 (doesn't match the Ground Truth's $91.4B).\n", "\n", "\n", "{\n", " \"faithfulness\": 1.0,\n", " \"relevance\": 1.0,\n", " \"correctness\": 0.0\n", "}\n", "For troubleshooting, visit: https://docs.langchain.com/oss/python/langchain/errors/OUTPUT_PARSING_FAILURE \n", "ERROR:financial_rag:Failed to parse LLM evaluation response: Invalid json output: \n", "Okay, let's start by looking at the user's question: they want to know how much Google spent on capital expenditures in fiscal year 2025. The context provided includes a lot of financial data from Google's filings. \n", "\n", "First, I need to check if the answer is present in the context. The user's answer states that the capital expenditures for 2025 aren't explicitly mentioned, but the Ground Truth says it's $91.4 billion. Let me scan through the context again.\n", "\n", "Looking through the sections, under \"Cash Used in Investing Activities\" for 2025, there's a line: \"$69.69 billion of purchases of property and equipment...\" Wait, property and equipment purchases are part of capital expenditures. But the Ground Truth mentions $91.4 billion. However, the context also says \"We anticipate making capital expenditures of approximately $115 billion to $135 billion in 2026...\" which is a future year. \n", "\n", "Wait, the user's answer says the 2025 capital expenditures aren't stated, but the Ground Truth claims it's $91.4B. Let me check again. The context under \"Cash Used in Investing Activities during 2025\" lists $69.69B for property and equipment. But capital expenditures can include more than just property and equipment. However, the Ground Truth says $91.4B, which isn't directly stated in the context. The context only mentions $69.69B for property and equipment. \n", "\n", "So the answer correctly points out that the context doesn't explicitly state the total capital expenditures for 2025. The Ground Truth is providing a figure that isn't in the context. Therefore, the answer is faithful because it doesn't make up a number not present. Relevance is high because it directly addresses the question. Correctness is low because the answer doesn't match the Ground Truth. The Ground Truth's $91.4B isn't in the context, so the answer is correct in stating it's not provided. Therefore, the correctness score should be 0.0 because the answer contradicts the Ground Truth by not providing the correct figure when it's actually present in the context? Wait, no. Wait, the context does mention $69.69B for property and equipment under 2025. But the Ground Truth says $91.4B. Is there another line item in the context that adds up to $91.4B? Let me check again.\n", "\n", "In the context under \"Cash Used in Investing Activities during 2025\": $69.69B (property and equipment), $18.33B (non-marketable equity investments), and $10.05B (marketable securities). Adding those gives $98.07B. But the Ground Truth says $91.4B. Hmm, maybe the Ground Truth is incorrect, or perhaps the answer is missing something. Alternatively, maybe the $69.69B is part of the total capital expenditures. But the answer says it's not stated, but the context does have a figure. Wait, the answer says the context doesn't provide a specific amount for capital expenditures, but the context does mention $69.69B for property and equipment. However, capital expenditures can include other items. The answer is correct in stating that the total isn't explicitly given, but the Ground Truth is providing a figure that isn't in the context. Therefore, the answer is faithful (since it doesn't hallucinate a number), relevant (addresses the question), but incorrect compared to the Ground Truth. However, the Ground Truth might be wrong if the context doesn't actually state $91.4B. Wait, the user's Ground Truth says $91.4B, but the context only mentions $69.69B for property and equipment. Unless there's another line item. Let me check again.\n", "\n", "Looking at the context: \"Cash used in investing activities during 2025 mostly consisted of $69.69 billion of purchases of property and equipment...\" So that's part of capital expenditures. But the total cash used in investing activities is $69.69B + $18.33B + $10.05B = $98.07B. However, capital expenditures typically refer to purchases of property and equipment. The other items (non-marketable equity investments and marketable securities) are not usually considered capital expenditures. Therefore, the actual capital expenditures for 2025 would be $69.69B. But the Ground Truth says $91.4B, which is not present in the context. Therefore, the answer is correct in stating that the context doesn't provide the total capital expenditures for 2025, and the Ground Truth is incorrect. However, the user's Ground Truth is given as a reference, so the answer's correctness is 0.0 because it doesn't match the Ground Truth. But wait, the answer is correct based on the context, but the Ground Truth is wrong. However, the user's instruction says to score correctness based on agreement with the Ground Truth. So even if the Ground Truth is incorrect, the answer's correctness is 0.0 because it doesn't match. Therefore, the answer is not correct compared to the Ground Truth. So the scores would be: faithfulness 1.0 (since it doesn't make up a number), relevance 1.0 (directly answers the question), correctness 0.0 (doesn't match the Ground Truth's $91.4B).\n", "\n", "\n", "{\n", " \"faithfulness\": 1.0,\n", " \"relevance\": 1.0,\n", " \"correctness\": 0.0\n", "}\n", "For troubleshooting, visit: https://docs.langchain.com/oss/python/langchain/errors/OUTPUT_PARSING_FAILURE \n", "2026-04-03 05:39:46 - [INFO] - financial_rag - Evaluating question 4/15: 'What was Google Cloud's revenue in fiscal year 2025?'\n", "INFO:financial_rag: Evaluating question 4/15: 'What was Google Cloud's revenue in fiscal year 2025?'\n", "2026-04-03 05:39:46 - [INFO] - financial_rag - Retrieving documents for query: 'What was Google Cloud's revenue in fiscal year 2025?'\n", "INFO:financial_rag:Retrieving documents for query: 'What was Google Cloud's revenue in fiscal year 2025?'\n", "2026-04-03 05:39:46 - [INFO] - financial_rag - Step 1: Executing Chain-of-Thought Analysis (llama-3.3-70b-versatile)...\n", "INFO:financial_rag:Step 1: Executing Chain-of-Thought Analysis (llama-3.3-70b-versatile)...\n", "2026-04-03 05:39:47 - [INFO] - financial_rag - Step 2: Running Strict Compliance Audit (llama-3.3-70b-versatile)...\n", "INFO:financial_rag:Step 2: Running Strict Compliance Audit (llama-3.3-70b-versatile)...\n", "2026-04-03 05:39:47 - [INFO] - financial_rag - Running LLM-as-a-Judge evaluation (Judge model: qwen/qwen3-32b)...\n", "INFO:financial_rag:Running LLM-as-a-Judge evaluation (Judge model: qwen/qwen3-32b)...\n", "2026-04-03 05:40:23 - [ERROR] - financial_rag - Failed to parse LLM evaluation response: Invalid json output: \n", "Okay, let's start by looking at the user's question: What was Google Cloud's revenue in fiscal year 2025? The answer provided is $58,705 million, citing the Google 10-K. The context given includes financial data tables where under the \"Revenues\" section, Google Cloud's 2025 revenue is listed as $58,705. So the answer is directly pulling that number from the context.\n", "\n", "First, checking faithfulness. The answer states the exact figure from the context without adding any other information. There's no mention of other sources or unsupported claims. So faithfulness should be 1.0.\n", "\n", "Next, relevance. The question asks specifically for Google Cloud's revenue in 2025, and the answer provides that exact number. It's directly addressing the question without any deviation. Relevance is 1.0.\n", "\n", "For correctness, the Ground Truth says the revenue was $58.705 billion, which is the same as $58,705 million. The answer matches this exactly. The Ground Truth also mentions a 36% increase, but the answer doesn't include that, but correctness here is about factual agreement. Since the key figure is correct, correctness is 1.0. Even though the Ground Truth has more details, the answer's main fact is correct.\n", "\n", "\n", "{\n", " \"faithfulness\": 1.0,\n", " \"relevance\": 1.0,\n", " \"correctness\": 1.0\n", "}\n", "For troubleshooting, visit: https://docs.langchain.com/oss/python/langchain/errors/OUTPUT_PARSING_FAILURE \n", "ERROR:financial_rag:Failed to parse LLM evaluation response: Invalid json output: \n", "Okay, let's start by looking at the user's question: What was Google Cloud's revenue in fiscal year 2025? The answer provided is $58,705 million, citing the Google 10-K. The context given includes financial data tables where under the \"Revenues\" section, Google Cloud's 2025 revenue is listed as $58,705. So the answer is directly pulling that number from the context.\n", "\n", "First, checking faithfulness. The answer states the exact figure from the context without adding any other information. There's no mention of other sources or unsupported claims. So faithfulness should be 1.0.\n", "\n", "Next, relevance. The question asks specifically for Google Cloud's revenue in 2025, and the answer provides that exact number. It's directly addressing the question without any deviation. Relevance is 1.0.\n", "\n", "For correctness, the Ground Truth says the revenue was $58.705 billion, which is the same as $58,705 million. The answer matches this exactly. The Ground Truth also mentions a 36% increase, but the answer doesn't include that, but correctness here is about factual agreement. Since the key figure is correct, correctness is 1.0. Even though the Ground Truth has more details, the answer's main fact is correct.\n", "\n", "\n", "{\n", " \"faithfulness\": 1.0,\n", " \"relevance\": 1.0,\n", " \"correctness\": 1.0\n", "}\n", "For troubleshooting, visit: https://docs.langchain.com/oss/python/langchain/errors/OUTPUT_PARSING_FAILURE \n", "2026-04-03 05:40:23 - [INFO] - financial_rag - Evaluating question 5/15: 'What were Meta's total revenues and net income in fiscal yea'\n", "INFO:financial_rag: Evaluating question 5/15: 'What were Meta's total revenues and net income in fiscal yea'\n", "2026-04-03 05:40:23 - [INFO] - financial_rag - Retrieving documents for query: 'What were Meta's total revenues and net income in fiscal year 2025?'\n", "INFO:financial_rag:Retrieving documents for query: 'What were Meta's total revenues and net income in fiscal year 2025?'\n", "2026-04-03 05:40:23 - [INFO] - financial_rag - Step 1: Executing Chain-of-Thought Analysis (llama-3.3-70b-versatile)...\n", "INFO:financial_rag:Step 1: Executing Chain-of-Thought Analysis (llama-3.3-70b-versatile)...\n", "2026-04-03 05:40:24 - [INFO] - financial_rag - Step 2: Running Strict Compliance Audit (llama-3.3-70b-versatile)...\n", "INFO:financial_rag:Step 2: Running Strict Compliance Audit (llama-3.3-70b-versatile)...\n", "2026-04-03 05:40:24 - [INFO] - financial_rag - Running LLM-as-a-Judge evaluation (Judge model: qwen/qwen3-32b)...\n", "INFO:financial_rag:Running LLM-as-a-Judge evaluation (Judge model: qwen/qwen3-32b)...\n", "2026-04-03 05:40:57 - [INFO] - financial_rag - Evaluation scores: {'faithfulness': 1.0, 'relevance': 1.0, 'correctness': 1.0}\n", "INFO:financial_rag:Evaluation scores: {'faithfulness': 1.0, 'relevance': 1.0, 'correctness': 1.0}\n", "2026-04-03 05:40:57 - [INFO] - financial_rag - Evaluating question 6/15: 'What were Meta's Research and Development expenses in fiscal'\n", "INFO:financial_rag: Evaluating question 6/15: 'What were Meta's Research and Development expenses in fiscal'\n", "2026-04-03 05:40:57 - [INFO] - financial_rag - Retrieving documents for query: 'What were Meta's Research and Development expenses in fiscal year 2025?'\n", "INFO:financial_rag:Retrieving documents for query: 'What were Meta's Research and Development expenses in fiscal year 2025?'\n", "2026-04-03 05:40:58 - [INFO] - financial_rag - Step 1: Executing Chain-of-Thought Analysis (llama-3.3-70b-versatile)...\n", "INFO:financial_rag:Step 1: Executing Chain-of-Thought Analysis (llama-3.3-70b-versatile)...\n", "2026-04-03 05:40:59 - [INFO] - financial_rag - Step 2: Running Strict Compliance Audit (llama-3.3-70b-versatile)...\n", "INFO:financial_rag:Step 2: Running Strict Compliance Audit (llama-3.3-70b-versatile)...\n", "2026-04-03 05:41:00 - [INFO] - financial_rag - Running LLM-as-a-Judge evaluation (Judge model: qwen/qwen3-32b)...\n", "INFO:financial_rag:Running LLM-as-a-Judge evaluation (Judge model: qwen/qwen3-32b)...\n", "2026-04-03 05:41:29 - [INFO] - financial_rag - Evaluation scores: {'faithfulness': 1.0, 'relevance': 1.0, 'correctness': 1.0}\n", "INFO:financial_rag:Evaluation scores: {'faithfulness': 1.0, 'relevance': 1.0, 'correctness': 1.0}\n", "2026-04-03 05:41:29 - [INFO] - financial_rag - Evaluating question 7/15: 'What was Meta's Reality Labs operating loss in 2025 and what'\n", "INFO:financial_rag: Evaluating question 7/15: 'What was Meta's Reality Labs operating loss in 2025 and what'\n", "2026-04-03 05:41:29 - [INFO] - financial_rag - Retrieving documents for query: 'What was Meta's Reality Labs operating loss in 2025 and what is the 2026 capital expenditure guidance?'\n", "INFO:financial_rag:Retrieving documents for query: 'What was Meta's Reality Labs operating loss in 2025 and what is the 2026 capital expenditure guidance?'\n", "2026-04-03 05:41:29 - [INFO] - financial_rag - Step 1: Executing Chain-of-Thought Analysis (llama-3.3-70b-versatile)...\n", "INFO:financial_rag:Step 1: Executing Chain-of-Thought Analysis (llama-3.3-70b-versatile)...\n", "2026-04-03 05:41:30 - [INFO] - financial_rag - Step 2: Running Strict Compliance Audit (llama-3.3-70b-versatile)...\n", "INFO:financial_rag:Step 2: Running Strict Compliance Audit (llama-3.3-70b-versatile)...\n", "2026-04-03 05:41:31 - [INFO] - financial_rag - Running LLM-as-a-Judge evaluation (Judge model: qwen/qwen3-32b)...\n", "INFO:financial_rag:Running LLM-as-a-Judge evaluation (Judge model: qwen/qwen3-32b)...\n", "2026-04-03 05:42:00 - [INFO] - financial_rag - Evaluation scores: {'faithfulness': 1.0, 'relevance': 1.0, 'correctness': 0.5}\n", "INFO:financial_rag:Evaluation scores: {'faithfulness': 1.0, 'relevance': 1.0, 'correctness': 0.5}\n", "2026-04-03 05:42:00 - [INFO] - financial_rag - Evaluating question 8/15: 'How many employees did Meta have as of December 31, 2025?'\n", "INFO:financial_rag: Evaluating question 8/15: 'How many employees did Meta have as of December 31, 2025?'\n", "2026-04-03 05:42:00 - [INFO] - financial_rag - Retrieving documents for query: 'How many employees did Meta have as of December 31, 2025?'\n", "INFO:financial_rag:Retrieving documents for query: 'How many employees did Meta have as of December 31, 2025?'\n", "2026-04-03 05:42:00 - [INFO] - financial_rag - Step 1: Executing Chain-of-Thought Analysis (llama-3.3-70b-versatile)...\n", "INFO:financial_rag:Step 1: Executing Chain-of-Thought Analysis (llama-3.3-70b-versatile)...\n", "2026-04-03 05:42:01 - [INFO] - financial_rag - Step 2: Running Strict Compliance Audit (llama-3.3-70b-versatile)...\n", "INFO:financial_rag:Step 2: Running Strict Compliance Audit (llama-3.3-70b-versatile)...\n", "2026-04-03 05:42:01 - [INFO] - financial_rag - Running LLM-as-a-Judge evaluation (Judge model: qwen/qwen3-32b)...\n", "INFO:financial_rag:Running LLM-as-a-Judge evaluation (Judge model: qwen/qwen3-32b)...\n", "2026-04-03 05:42:20 - [INFO] - financial_rag - Evaluation scores: {'faithfulness': 1.0, 'relevance': 1.0, 'correctness': 1.0}\n", "INFO:financial_rag:Evaluation scores: {'faithfulness': 1.0, 'relevance': 1.0, 'correctness': 1.0}\n", "2026-04-03 05:42:20 - [INFO] - financial_rag - Evaluating question 9/15: 'What were Microsoft's total revenues and net income in fisca'\n", "INFO:financial_rag: Evaluating question 9/15: 'What were Microsoft's total revenues and net income in fisca'\n", "2026-04-03 05:42:20 - [INFO] - financial_rag - Retrieving documents for query: 'What were Microsoft's total revenues and net income in fiscal year 2024?'\n", "INFO:financial_rag:Retrieving documents for query: 'What were Microsoft's total revenues and net income in fiscal year 2024?'\n", "2026-04-03 05:42:20 - [INFO] - financial_rag - Step 1: Executing Chain-of-Thought Analysis (llama-3.3-70b-versatile)...\n", "INFO:financial_rag:Step 1: Executing Chain-of-Thought Analysis (llama-3.3-70b-versatile)...\n", "2026-04-03 05:42:21 - [INFO] - financial_rag - Step 2: Running Strict Compliance Audit (llama-3.3-70b-versatile)...\n", "INFO:financial_rag:Step 2: Running Strict Compliance Audit (llama-3.3-70b-versatile)...\n", "2026-04-03 05:42:22 - [INFO] - financial_rag - Running LLM-as-a-Judge evaluation (Judge model: qwen/qwen3-32b)...\n", "INFO:financial_rag:Running LLM-as-a-Judge evaluation (Judge model: qwen/qwen3-32b)...\n", "2026-04-03 05:42:57 - [ERROR] - financial_rag - Failed to parse LLM evaluation response: Invalid json output: \n", "Okay, let's start by looking at the user's question: they want to know Microsoft's total revenues and net income in fiscal year 2024. The context provided includes financial data from Microsoft's 10-K report. \n", "\n", "First, I need to check the Answer against the Context for faithfulness. The Answer states that total revenue was $245,122 million, which is listed under the \"Segment Results of Operations\" in the Context. That's correct. However, the Answer mentions that net income isn't provided in the context. Looking at the Context, there's a line that says \"Net income $100,118 $132,170 $32,052 32%\", which shows the net income for 2024 as $100,118 million. The Answer didn't include this, so it's missing a key piece of information. Therefore, the faithfulness score should be lower because the Answer doesn't include the net income from the Context.\n", "\n", "Next, relevance. The Answer addresses the total revenue but doesn't mention the net income, which the user asked for. Since the user specifically asked for both, the Answer is incomplete. However, it does correctly state that the net income isn't provided in the context, which is accurate because the Context does have the net income figure. Wait, the Context does have the net income for 2024 as $100,118 million. The Answer says it's not provided, which is incorrect. So the Answer is not relevant because it fails to provide the net income when it's actually present in the Context. Therefore, the relevance score should be low.\n", "\n", "For correctness, the Ground Truth states total revenue as $245.122 billion and net income as $88.136 billion. The Answer correctly reports the total revenue from the Context. However, the Ground Truth's net income is $88.136 billion, but the Context shows $100,118 million (which is $100.118 billion). There's a discrepancy here. The Answer didn't include the net income, so it's not correct. The Ground Truth's net income doesn't match the Context's figure. This might be a problem. Wait, the user's Ground Truth says $88.136 billion, but the Context shows $100,118 million. That's a significant difference. The Answer didn't provide the net income, so it's not correct. Therefore, the correctness score is low because the Answer doesn't match the Ground Truth's net income, and the Ground Truth itself might be conflicting with the Context. But since the Answer didn't include the net income, it's not correct. So correctness is 0.5? Wait, the Answer didn't provide the net income, so it's missing a key fact. The Ground Truth has a different number than the Context. The Answer didn't include the net income, so it's not correct. Therefore, correctness is 0.0 because it's missing the net income and the Ground Truth's figure is different from the Context. But the user's Ground Truth might be incorrect. However, according to the instructions, we have to compare the Answer to the Ground Truth. The Answer didn't provide the net income, so it's incorrect. So correctness is 0.0.\n", "\n", "\n", "{\n", " \"faithfulness\": 0.5,\n", " \"relevance\": 0.5,\n", " \"correctness\": 0.0\n", "}\n", "For troubleshooting, visit: https://docs.langchain.com/oss/python/langchain/errors/OUTPUT_PARSING_FAILURE \n", "ERROR:financial_rag:Failed to parse LLM evaluation response: Invalid json output: \n", "Okay, let's start by looking at the user's question: they want to know Microsoft's total revenues and net income in fiscal year 2024. The context provided includes financial data from Microsoft's 10-K report. \n", "\n", "First, I need to check the Answer against the Context for faithfulness. The Answer states that total revenue was $245,122 million, which is listed under the \"Segment Results of Operations\" in the Context. That's correct. However, the Answer mentions that net income isn't provided in the context. Looking at the Context, there's a line that says \"Net income $100,118 $132,170 $32,052 32%\", which shows the net income for 2024 as $100,118 million. The Answer didn't include this, so it's missing a key piece of information. Therefore, the faithfulness score should be lower because the Answer doesn't include the net income from the Context.\n", "\n", "Next, relevance. The Answer addresses the total revenue but doesn't mention the net income, which the user asked for. Since the user specifically asked for both, the Answer is incomplete. However, it does correctly state that the net income isn't provided in the context, which is accurate because the Context does have the net income figure. Wait, the Context does have the net income for 2024 as $100,118 million. The Answer says it's not provided, which is incorrect. So the Answer is not relevant because it fails to provide the net income when it's actually present in the Context. Therefore, the relevance score should be low.\n", "\n", "For correctness, the Ground Truth states total revenue as $245.122 billion and net income as $88.136 billion. The Answer correctly reports the total revenue from the Context. However, the Ground Truth's net income is $88.136 billion, but the Context shows $100,118 million (which is $100.118 billion). There's a discrepancy here. The Answer didn't include the net income, so it's not correct. The Ground Truth's net income doesn't match the Context's figure. This might be a problem. Wait, the user's Ground Truth says $88.136 billion, but the Context shows $100,118 million. That's a significant difference. The Answer didn't provide the net income, so it's not correct. Therefore, the correctness score is low because the Answer doesn't match the Ground Truth's net income, and the Ground Truth itself might be conflicting with the Context. But since the Answer didn't include the net income, it's not correct. So correctness is 0.5? Wait, the Answer didn't provide the net income, so it's missing a key fact. The Ground Truth has a different number than the Context. The Answer didn't include the net income, so it's not correct. Therefore, correctness is 0.0 because it's missing the net income and the Ground Truth's figure is different from the Context. But the user's Ground Truth might be incorrect. However, according to the instructions, we have to compare the Answer to the Ground Truth. The Answer didn't provide the net income, so it's incorrect. So correctness is 0.0.\n", "\n", "\n", "{\n", " \"faithfulness\": 0.5,\n", " \"relevance\": 0.5,\n", " \"correctness\": 0.0\n", "}\n", "For troubleshooting, visit: https://docs.langchain.com/oss/python/langchain/errors/OUTPUT_PARSING_FAILURE \n", "2026-04-03 05:42:57 - [INFO] - financial_rag - Evaluating question 10/15: 'What was Microsoft's cloud revenue and what is Microsoft's p'\n", "INFO:financial_rag: Evaluating question 10/15: 'What was Microsoft's cloud revenue and what is Microsoft's p'\n", "2026-04-03 05:42:57 - [INFO] - financial_rag - Retrieving documents for query: 'What was Microsoft's cloud revenue and what is Microsoft's primary cloud platform?'\n", "INFO:financial_rag:Retrieving documents for query: 'What was Microsoft's cloud revenue and what is Microsoft's primary cloud platform?'\n", "2026-04-03 05:42:57 - [INFO] - financial_rag - Step 1: Executing Chain-of-Thought Analysis (llama-3.3-70b-versatile)...\n", "INFO:financial_rag:Step 1: Executing Chain-of-Thought Analysis (llama-3.3-70b-versatile)...\n", "2026-04-03 05:42:58 - [INFO] - financial_rag - Step 2: Running Strict Compliance Audit (llama-3.3-70b-versatile)...\n", "INFO:financial_rag:Step 2: Running Strict Compliance Audit (llama-3.3-70b-versatile)...\n", "2026-04-03 05:42:59 - [INFO] - financial_rag - Running LLM-as-a-Judge evaluation (Judge model: qwen/qwen3-32b)...\n", "INFO:financial_rag:Running LLM-as-a-Judge evaluation (Judge model: qwen/qwen3-32b)...\n", "2026-04-03 05:43:31 - [INFO] - financial_rag - Evaluation scores: {'faithfulness': 0.5, 'relevance': 1.0, 'correctness': 1.0}\n", "INFO:financial_rag:Evaluation scores: {'faithfulness': 0.5, 'relevance': 1.0, 'correctness': 1.0}\n", "2026-04-03 05:43:31 - [INFO] - financial_rag - Evaluating question 11/15: 'Compare the R&D spending of Google, Meta, and Microsoft in t'\n", "INFO:financial_rag: Evaluating question 11/15: 'Compare the R&D spending of Google, Meta, and Microsoft in t'\n", "2026-04-03 05:43:31 - [INFO] - financial_rag - Retrieving documents for query: 'Compare the R&D spending of Google, Meta, and Microsoft in their most recent fiscal years.'\n", "INFO:financial_rag:Retrieving documents for query: 'Compare the R&D spending of Google, Meta, and Microsoft in their most recent fiscal years.'\n", "2026-04-03 05:43:31 - [INFO] - financial_rag - Step 1: Executing Chain-of-Thought Analysis (llama-3.3-70b-versatile)...\n", "INFO:financial_rag:Step 1: Executing Chain-of-Thought Analysis (llama-3.3-70b-versatile)...\n", "2026-04-03 05:43:33 - [INFO] - financial_rag - Step 2: Running Strict Compliance Audit (llama-3.3-70b-versatile)...\n", "INFO:financial_rag:Step 2: Running Strict Compliance Audit (llama-3.3-70b-versatile)...\n", "2026-04-03 05:43:34 - [INFO] - financial_rag - Running LLM-as-a-Judge evaluation (Judge model: qwen/qwen3-32b)...\n", "INFO:financial_rag:Running LLM-as-a-Judge evaluation (Judge model: qwen/qwen3-32b)...\n", "2026-04-03 05:44:04 - [INFO] - financial_rag - Evaluation scores: {'faithfulness': 0.0, 'relevance': 0.5, 'correctness': 0.0}\n", "INFO:financial_rag:Evaluation scores: {'faithfulness': 0.0, 'relevance': 0.5, 'correctness': 0.0}\n", "2026-04-03 05:44:04 - [INFO] - financial_rag - Evaluating question 12/15: 'Compare the capital expenditures of Google and Meta in their'\n", "INFO:financial_rag: Evaluating question 12/15: 'Compare the capital expenditures of Google and Meta in their'\n", "2026-04-03 05:44:04 - [INFO] - financial_rag - Retrieving documents for query: 'Compare the capital expenditures of Google and Meta in their most recent fiscal years and explain what they are investing in.'\n", "INFO:financial_rag:Retrieving documents for query: 'Compare the capital expenditures of Google and Meta in their most recent fiscal years and explain what they are investing in.'\n", "2026-04-03 05:44:04 - [INFO] - financial_rag - Step 1: Executing Chain-of-Thought Analysis (llama-3.3-70b-versatile)...\n", "INFO:financial_rag:Step 1: Executing Chain-of-Thought Analysis (llama-3.3-70b-versatile)...\n", "2026-04-03 05:44:07 - [INFO] - financial_rag - Step 2: Running Strict Compliance Audit (llama-3.3-70b-versatile)...\n", "INFO:financial_rag:Step 2: Running Strict Compliance Audit (llama-3.3-70b-versatile)...\n", "2026-04-03 05:44:08 - [INFO] - financial_rag - Running LLM-as-a-Judge evaluation (Judge model: qwen/qwen3-32b)...\n", "INFO:financial_rag:Running LLM-as-a-Judge evaluation (Judge model: qwen/qwen3-32b)...\n", "2026-04-03 05:44:30 - [INFO] - financial_rag - Evaluation scores: {'faithfulness': 1.0, 'relevance': 1.0, 'correctness': 1.0}\n", "INFO:financial_rag:Evaluation scores: {'faithfulness': 1.0, 'relevance': 1.0, 'correctness': 1.0}\n", "2026-04-03 05:44:30 - [INFO] - financial_rag - Evaluating question 13/15: 'What are the primary regulatory and legal risks mentioned by'\n", "INFO:financial_rag: Evaluating question 13/15: 'What are the primary regulatory and legal risks mentioned by'\n", "2026-04-03 05:44:30 - [INFO] - financial_rag - Retrieving documents for query: 'What are the primary regulatory and legal risks mentioned by Google in its 10-K filing?'\n", "INFO:financial_rag:Retrieving documents for query: 'What are the primary regulatory and legal risks mentioned by Google in its 10-K filing?'\n", "2026-04-03 05:44:30 - [INFO] - financial_rag - Step 1: Executing Chain-of-Thought Analysis (llama-3.3-70b-versatile)...\n", "INFO:financial_rag:Step 1: Executing Chain-of-Thought Analysis (llama-3.3-70b-versatile)...\n", "2026-04-03 05:44:32 - [INFO] - financial_rag - Step 2: Running Strict Compliance Audit (llama-3.3-70b-versatile)...\n", "INFO:financial_rag:Step 2: Running Strict Compliance Audit (llama-3.3-70b-versatile)...\n", "2026-04-03 05:44:33 - [INFO] - financial_rag - Running LLM-as-a-Judge evaluation (Judge model: qwen/qwen3-32b)...\n", "INFO:financial_rag:Running LLM-as-a-Judge evaluation (Judge model: qwen/qwen3-32b)...\n", "2026-04-03 05:44:57 - [INFO] - financial_rag - Evaluation scores: {'faithfulness': 1.0, 'relevance': 1.0}\n", "INFO:financial_rag:Evaluation scores: {'faithfulness': 1.0, 'relevance': 1.0}\n", "2026-04-03 05:44:57 - [INFO] - financial_rag - Evaluating question 14/15: 'What are the main risk factors Meta identifies for its adver'\n", "INFO:financial_rag: Evaluating question 14/15: 'What are the main risk factors Meta identifies for its adver'\n", "2026-04-03 05:44:57 - [INFO] - financial_rag - Retrieving documents for query: 'What are the main risk factors Meta identifies for its advertising business in its 10-K?'\n", "INFO:financial_rag:Retrieving documents for query: 'What are the main risk factors Meta identifies for its advertising business in its 10-K?'\n", "2026-04-03 05:44:57 - [INFO] - financial_rag - Step 1: Executing Chain-of-Thought Analysis (llama-3.3-70b-versatile)...\n", "INFO:financial_rag:Step 1: Executing Chain-of-Thought Analysis (llama-3.3-70b-versatile)...\n", "2026-04-03 05:44:58 - [INFO] - financial_rag - Step 2: Running Strict Compliance Audit (llama-3.3-70b-versatile)...\n", "INFO:financial_rag:Step 2: Running Strict Compliance Audit (llama-3.3-70b-versatile)...\n", "2026-04-03 05:44:59 - [INFO] - financial_rag - Running LLM-as-a-Judge evaluation (Judge model: qwen/qwen3-32b)...\n", "INFO:financial_rag:Running LLM-as-a-Judge evaluation (Judge model: qwen/qwen3-32b)...\n", "2026-04-03 05:45:23 - [INFO] - financial_rag - Evaluation scores: {'faithfulness': 1.0, 'relevance': 1.0}\n", "INFO:financial_rag:Evaluation scores: {'faithfulness': 1.0, 'relevance': 1.0}\n", "2026-04-03 05:45:23 - [INFO] - financial_rag - Evaluating question 15/15: 'How does Microsoft describe its AI strategy and investments '\n", "INFO:financial_rag: Evaluating question 15/15: 'How does Microsoft describe its AI strategy and investments '\n", "2026-04-03 05:45:23 - [INFO] - financial_rag - Retrieving documents for query: 'How does Microsoft describe its AI strategy and investments in its 10-K filing?'\n", "INFO:financial_rag:Retrieving documents for query: 'How does Microsoft describe its AI strategy and investments in its 10-K filing?'\n", "2026-04-03 05:45:23 - [INFO] - financial_rag - Step 1: Executing Chain-of-Thought Analysis (llama-3.3-70b-versatile)...\n", "INFO:financial_rag:Step 1: Executing Chain-of-Thought Analysis (llama-3.3-70b-versatile)...\n", "2026-04-03 05:45:25 - [INFO] - financial_rag - Step 2: Running Strict Compliance Audit (llama-3.3-70b-versatile)...\n", "INFO:financial_rag:Step 2: Running Strict Compliance Audit (llama-3.3-70b-versatile)...\n", "2026-04-03 05:45:26 - [INFO] - financial_rag - Running LLM-as-a-Judge evaluation (Judge model: qwen/qwen3-32b)...\n", "INFO:financial_rag:Running LLM-as-a-Judge evaluation (Judge model: qwen/qwen3-32b)...\n", "2026-04-03 05:45:56 - [INFO] - financial_rag - Evaluation scores: {'faithfulness': 1.0, 'relevance': 1.0}\n", "INFO:financial_rag:Evaluation scores: {'faithfulness': 1.0, 'relevance': 1.0}\n", "2026-04-03 05:45:56 - [INFO] - financial_rag - Batch evaluation complete. Faithfulness: 0.864 ± 0.323 | Relevance: 0.955 ± 0.151 | Pass rate: 81.8% | n=15\n", "INFO:financial_rag:Batch evaluation complete. Faithfulness: 0.864 ± 0.323 | Relevance: 0.955 ± 0.151 | Pass rate: 81.8% | n=15\n", "2026-04-03 05:45:56 - [INFO] - financial_rag - Batch evaluation report saved to: /content/drive/MyDrive/financial-intelligence-engine/artifacts/eval_reports/batch_eval_report.json\n", "INFO:financial_rag:Batch evaluation report saved to: /content/drive/MyDrive/financial-intelligence-engine/artifacts/eval_reports/batch_eval_report.json\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\n", "=================================================================\n", " BATCH EVALUATION RESULTS\n", "=================================================================\n", " Metric Mean Std Dev Pass Rate\n", " Faithfulness (No Hallucinations) 0.864 ±0.323 81.8%\n", " Context Relevance (Answers Prompt) 0.955 ±0.151 N/A\n", "Answer Correctness (vs Ground Truth) 0.812 ±0.372 N/A\n", "\n", "Total questions evaluated : 15\n", "Generator model : llama-3.3-70b-versatile\n", "Evaluator model : qwen/qwen3-32b\n", "Report saved to : artifacts/eval_reports/batch_eval_report.json\n", "\n", "=================================================================\n" ] } ], "source": [ "# ============================================================\n", "# Cell 5 — Batch Evaluation (Statistically Valid)\n", "# ============================================================\n", "# All ground_truth values are extracted DIRECTLY from the\n", "# uploaded 10-K PDF filings. Every number is verified.\n", "#\n", "# Fiscal Year Note:\n", "# Google: FY2025 (Jan 1 – Dec 31, 2025)\n", "# Meta: FY2025 (Jan 1 – Dec 31, 2025)\n", "# Microsoft: FY2024 (Jul 1, 2023 – Jun 30, 2024)\n", "# ============================================================\n", "\n", "import pandas as pd\n", "from src.evaluation import RAGEvaluator\n", "\n", "# ── Verified Evaluation Set ───────────────────────────────────\n", "# ground_truth strings are exact figures from the 10-K filings.\n", "# Questions WITHOUT ground_truth=None test faithfulness/relevance only.\n", "# Questions WITH ground_truth test all 3 metrics including correctness.\n", "\n", "EVAL_QUESTIONS = [\n", "\n", " # ── Google Questions ──────────────────────────────────────\n", " {\n", " \"question\": \"What were Google's total Research and Development expenses in fiscal year 2025?\",\n", " \"ground_truth\": (\n", " \"Google's Research and Development expenses were $61.087 billion \"\n", " \"in fiscal year 2025, representing 15% of total revenues. \"\n", " \"This was an increase of $11.8 billion compared to 2024.\"\n", " )\n", " },\n", " {\n", " \"question\": \"What were Google's total revenues and net income in fiscal year 2025?\",\n", " \"ground_truth\": (\n", " \"Google reported total revenues of $402.836 billion and net income \"\n", " \"of $132.170 billion in fiscal year 2025. Revenue increased 15% year over year.\"\n", " )\n", " },\n", " {\n", " \"question\": \"How much did Google spend on capital expenditures in fiscal year 2025?\",\n", " \"ground_truth\": (\n", " \"Google spent $91.4 billion on capital expenditures in fiscal year 2025, \"\n", " \"primarily reflecting investments in technical infrastructure including \"\n", " \"servers, network equipment, and data centers.\"\n", " )\n", " },\n", " {\n", " \"question\": \"What was Google Cloud's revenue in fiscal year 2025?\",\n", " \"ground_truth\": (\n", " \"Google Cloud revenue was $58.705 billion in fiscal year 2025, \"\n", " \"an increase of $15.5 billion or 36% compared to 2024.\"\n", " )\n", " },\n", "\n", " # ── Meta Questions ────────────────────────────────────────\n", " {\n", " \"question\": \"What were Meta's total revenues and net income in fiscal year 2025?\",\n", " \"ground_truth\": (\n", " \"Meta reported total revenue of $200.966 billion and net income of \"\n", " \"$60.458 billion in fiscal year 2025. Revenue increased 22% compared to 2024.\"\n", " )\n", " },\n", " {\n", " \"question\": \"What were Meta's Research and Development expenses in fiscal year 2025?\",\n", " \"ground_truth\": (\n", " \"Meta's Research and Development expenses were $57.372 billion in fiscal year 2025, \"\n", " \"representing 29% of total revenue. This was an increase of $13.5 billion or 31% \"\n", " \"compared to 2024.\"\n", " )\n", " },\n", " {\n", " \"question\": \"What was Meta's Reality Labs operating loss in 2025 and what is the 2026 capital expenditure guidance?\",\n", " \"ground_truth\": (\n", " \"Meta's Reality Labs segment reduced overall operating profit by approximately \"\n", " \"$19.19 billion in 2025. Meta anticipates capital expenditures of approximately \"\n", " \"$115 billion to $135 billion in 2026 to support AI efforts and core business.\"\n", " )\n", " },\n", " {\n", " \"question\": \"How many employees did Meta have as of December 31, 2025?\",\n", " \"ground_truth\": (\n", " \"Meta had a global workforce of 78,865 employees as of December 31, 2025, \"\n", " \"with offices in more than 90 cities around the world.\"\n", " )\n", " },\n", "\n", " # ── Microsoft Questions ───────────────────────────────────\n", " {\n", " \"question\": \"What were Microsoft's total revenues and net income in fiscal year 2024?\",\n", " \"ground_truth\": (\n", " \"Microsoft reported total revenue of $245.122 billion and net income of \"\n", " \"$88.136 billion in fiscal year 2024, which ended June 30, 2024. \"\n", " \"Revenue increased $33.2 billion or 16% year over year.\"\n", " )\n", " },\n", " {\n", " \"question\": \"What was Microsoft's cloud revenue and what is Microsoft's primary cloud platform?\",\n", " \"ground_truth\": (\n", " \"Microsoft Cloud revenue was $137.4 billion in fiscal year 2024. \"\n", " \"Microsoft Azure is the primary cloud platform and falls under the \"\n", " \"Intelligent Cloud segment, which generated $105.362 billion in revenue. \"\n", " \"Azure and other cloud services revenue grew 30% year over year.\"\n", " )\n", " },\n", "\n", " # ── Cross-Company Comparison Questions ───────────────────\n", " # These have no single ground truth — they test faithfulness and relevance.\n", " {\n", " \"question\": \"Compare the R&D spending of Google, Meta, and Microsoft in their most recent fiscal years.\",\n", " \"ground_truth\": (\n", " \"Google spent $61.087 billion on R&D in FY2025. Meta spent $57.372 billion \"\n", " \"on R&D in FY2025. Microsoft spent $29.510 billion on R&D in FY2024. \"\n", " \"Google and Meta both invested roughly 15% and 29% of revenue respectively, \"\n", " \"while Microsoft invested 12% of revenue in R&D.\"\n", " )\n", " },\n", " {\n", " \"question\": \"Compare the capital expenditures of Google and Meta in their most recent fiscal years and explain what they are investing in.\",\n", " \"ground_truth\": (\n", " \"Google spent $91.4 billion on capital expenditures in FY2025, primarily \"\n", " \"in technical infrastructure including servers, network equipment, and data centers. \"\n", " \"Meta spent $72.22 billion on capital expenditures in FY2025 to support AI efforts \"\n", " \"and core business operations.\"\n", " )\n", " },\n", " {\n", " \"question\": \"What are the primary regulatory and legal risks mentioned by Google in its 10-K filing?\",\n", " \"ground_truth\": None # open-ended qualitative — tests faithfulness only\n", " },\n", " {\n", " \"question\": \"What are the main risk factors Meta identifies for its advertising business in its 10-K?\",\n", " \"ground_truth\": None # open-ended qualitative — tests faithfulness only\n", " },\n", " {\n", " \"question\": \"How does Microsoft describe its AI strategy and investments in its 10-K filing?\",\n", " \"ground_truth\": None # open-ended qualitative — tests faithfulness only\n", " },\n", "]\n", "\n", "# ── Run Batch Evaluation ──────────────────────────────────────\n", "print(f\"Phase 4b: Running Batch Evaluation over {len(EVAL_QUESTIONS)} questions...\")\n", "print(f\"Judge model : {evaluator.llm.model_name}\")\n", "print(f\"Generator : {agent.llm.model_name}\")\n", "print(\"This will take 5-8 minutes. Each question runs full generation + evaluation.\\n\")\n", "\n", "batch_results = evaluator.run_batch_evaluation(\n", " eval_set=EVAL_QUESTIONS,\n", " agent=agent,\n", " save_report=True,\n", ")\n", "\n", "# ── Print Aggregate Results Table ─────────────────────────────\n", "print(\"\\n\" + \"=\" * 65)\n", "print(\" BATCH EVALUATION RESULTS\")\n", "print(\"=\" * 65)\n", "\n", "summary_data = [\n", " {\n", " \"Metric\": \"Faithfulness (No Hallucinations)\",\n", " \"Mean\": f\"{batch_results['mean_faithfulness']:.3f}\",\n", " \"Std Dev\": f\"±{batch_results['std_faithfulness']:.3f}\",\n", " \"Pass Rate\": f\"{batch_results['faithfulness_pass_rate']*100:.1f}%\",\n", " },\n", " {\n", " \"Metric\": \"Context Relevance (Answers Prompt)\",\n", " \"Mean\": f\"{batch_results['mean_relevance']:.3f}\",\n", " \"Std Dev\": f\"±{batch_results['std_relevance']:.3f}\",\n", " \"Pass Rate\": \"N/A\",\n", " },\n", "]\n", "\n", "if \"mean_correctness\" in batch_results:\n", " summary_data.append({\n", " \"Metric\": \"Answer Correctness (vs Ground Truth)\",\n", " \"Mean\": f\"{batch_results['mean_correctness']:.3f}\",\n", " \"Std Dev\": f\"±{batch_results['std_correctness']:.3f}\",\n", " \"Pass Rate\": \"N/A\",\n", " })\n", "\n", "summary_df = pd.DataFrame(summary_data)\n", "print(summary_df.to_string(index=False))\n", "print(f\"\\nTotal questions evaluated : {batch_results['n']}\")\n", "print(f\"Generator model : {agent.llm.model_name}\")\n", "print(f\"Evaluator model : {batch_results['evaluator_model']}\")\n", "print(f\"Report saved to : artifacts/eval_reports/batch_eval_report.json\")\n", "print(\"\\n\" + \"=\" * 65)" ] }, { "cell_type": "code", "execution_count": 12, "id": "c2dda694", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Phase 5: Visualizing Engine Telemetry...\n", "Primary dashboard saved: /content/drive/MyDrive/financial-intelligence-engine/artifacts/visualizations/batch_eval_primary.png\n" ] }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "Telemetry dashboard saved: /content/drive/MyDrive/financial-intelligence-engine/artifacts/visualizations/telemetry_dashboard.png\n" ] }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "\n", "=================================================================\n", " FINANCIAL INTELLIGENCE ENGINE — FINAL METRICS SUMMARY\n", "=================================================================\n", " Faithfulness (n=15) : 0.864 ± 0.323 | Pass rate: 81.8%\n", " Relevance (n=15) : 0.955 ± 0.151\n", " Correctness (n=15) : 0.812 ± 0.372\n", " Generator model : llama-3.3-70b-versatile\n", " Evaluator model : qwen/qwen3-32b\n", " Retrieval method : Hybrid RRF (Dense ChromaDB + Sparse BM25)\n", " Company balance : {'Meta': 3, 'Microsoft': 2, 'Google': 2}\n", "=================================================================\n", "All artifacts saved to Google Drive /artifacts/ directory.\n" ] } ], "source": [ "# ============================================================\n", "# Cell 6 — Enterprise Visualization & Reporting\n", "# ============================================================\n", "# UPGRADE: Batch evaluation results are now the PRIMARY dashboard.\n", "# Single-query scores are shown as a secondary reference only.\n", "# This ensures the statistically valid n=15 scores are the first\n", "# thing a recruiter or interviewer sees — not a single noisy sample.\n", "# ============================================================\n", "\n", "import pandas as pd\n", "import matplotlib.pyplot as plt\n", "import matplotlib.patches as mpatches\n", "import seaborn as sns\n", "from datetime import datetime\n", "from collections import Counter\n", "from src.config import VISUALS_DIR\n", "\n", "print(\"Phase 5: Visualizing Engine Telemetry...\")\n", "\n", "sns.set_theme(style=\"whitegrid\", context=\"talk\")\n", "\n", "# ── DASHBOARD 1: Batch Evaluation Results (PRIMARY) ───────────\n", "# This is the statistically valid result — shown first and largest.\n", "fig1, axes1 = plt.subplots(1, 2, figsize=(16, 7))\n", "fig1.suptitle(\n", " 'Financial Intelligence Engine — Batch Evaluation Dashboard\\n'\n", " f'n={batch_results[\"n\"]} Questions | Generator: {agent.llm.model_name} | '\n", " f'Judge: {batch_results[\"evaluator_model\"]}',\n", " fontsize=14, fontweight='bold', y=1.03\n", ")\n", "\n", "# ── Plot 1A: Batch Metric Scores with Error Bars ──────────────\n", "batch_metrics = ['Faithfulness', 'Relevance']\n", "batch_means = [batch_results['mean_faithfulness'], batch_results['mean_relevance']]\n", "batch_stds = [batch_results['std_faithfulness'], batch_results['std_relevance']]\n", "\n", "if 'mean_correctness' in batch_results:\n", " batch_metrics.append('Correctness\\n(vs Ground Truth)')\n", " batch_means.append(batch_results['mean_correctness'])\n", " batch_stds.append(batch_results['std_correctness'])\n", "\n", "bar_colors_batch = ['#2ecc71' if m >= 0.8 else '#e74c3c' for m in batch_means]\n", "x_pos = range(len(batch_metrics))\n", "\n", "bars = axes1[0].bar(\n", " x_pos, batch_means, yerr=batch_stds, capsize=7,\n", " color=bar_colors_batch, alpha=0.88, width=0.5,\n", " error_kw={'elinewidth': 2, 'capthick': 2}\n", ")\n", "axes1[0].set_xticks(list(x_pos))\n", "axes1[0].set_xticklabels(batch_metrics, fontsize=11)\n", "axes1[0].set_ylim(0, 1.2)\n", "axes1[0].set_ylabel('Mean Score (0.0 – 1.0)', fontweight='bold')\n", "axes1[0].set_title(\n", " f'LLM-as-a-Judge Scores (n={batch_results[\"n\"]})',\n", " fontsize=13, pad=12\n", ")\n", "axes1[0].axhline(\n", " y=0.8, color='gray', linestyle='--', linewidth=1.2,\n", " alpha=0.7, label='Pass threshold (0.80)'\n", ")\n", "axes1[0].legend(fontsize=10)\n", "\n", "# Value labels above bars\n", "for i, (m, s) in enumerate(zip(batch_means, batch_stds)):\n", " axes1[0].text(\n", " i, m + s + 0.04, f'{m:.3f}',\n", " ha='center', fontweight='bold', fontsize=12\n", " )\n", "\n", "# Pass rate annotation\n", "pass_rate = batch_results['faithfulness_pass_rate'] * 100\n", "axes1[0].text(\n", " 0.5, 0.05,\n", " f'Faithfulness Pass Rate: {pass_rate:.1f}%',\n", " transform=axes1[0].transAxes,\n", " ha='center', fontsize=10,\n", " color='#27ae60' if pass_rate >= 80 else '#e74c3c',\n", " fontweight='bold'\n", ")\n", "\n", "# ── Plot 1B: Retrieved Context Distribution ───────────────────\n", "source_counts = Counter(d.metadata.get('company', 'Unknown') for d in used_docs)\n", "labels = list(source_counts.keys())\n", "sizes = list(source_counts.values())\n", "explode = [0.05] * len(labels)\n", "\n", "axes1[1].pie(\n", " sizes,\n", " labels=labels,\n", " autopct='%1.1f%%',\n", " startangle=140,\n", " colors=sns.color_palette('deep', len(labels)),\n", " explode=explode,\n", " shadow=True,\n", " textprops={'fontsize': 12, 'fontweight': 'bold'},\n", ")\n", "axes1[1].set_title(\n", " 'Retrieved Context Distribution by Company\\n(Company-Balanced RRF)',\n", " fontsize=13, pad=15\n", ")\n", "\n", "plt.tight_layout()\n", "batch_primary_path = f'{VISUALS_DIR}/batch_eval_primary.png'\n", "plt.savefig(batch_primary_path, dpi=300, bbox_inches='tight')\n", "print(f'Primary dashboard saved: {batch_primary_path}')\n", "plt.show()\n", "\n", "\n", "# ── DASHBOARD 2: System Telemetry (SECONDARY) ─────────────────\n", "# Shows single-query result alongside batch summary for comparison.\n", "fig2, axes2 = plt.subplots(1, 2, figsize=(16, 7))\n", "fig2.suptitle(\n", " 'Financial Intelligence Engine — System Telemetry',\n", " fontsize=16, fontweight='bold', y=1.02\n", ")\n", "\n", "# ── Plot 2A: Single-Query Scores ──────────────────────────────\n", "metric_labels = [\n", " 'Faithfulness\\n(No Hallucinations)',\n", " 'Context Relevance\\n(Answers Prompt)'\n", "]\n", "score_vals = [\n", " float(scores.get('faithfulness', 0)),\n", " float(scores.get('relevance', 0)),\n", "]\n", "bar_colors_single = ['#2ecc71' if s >= 0.8 else '#e74c3c' for s in score_vals]\n", "\n", "plot_df = pd.DataFrame({\n", " 'Metric': metric_labels,\n", " 'Score': score_vals,\n", " 'Color': bar_colors_single,\n", "})\n", "sns.barplot(\n", " data=plot_df,\n", " x='Metric', y='Score',\n", " hue='Metric',\n", " palette=dict(zip(metric_labels, bar_colors_single)),\n", " legend=False,\n", " ax=axes2[0],\n", ")\n", "axes2[0].set_ylim(0, 1.15)\n", "axes2[0].set_title(\n", " 'Single-Query Validation\\n(Directional — see batch results for statistical validity)',\n", " fontsize=12, pad=12\n", ")\n", "axes2[0].set_ylabel('Score (0.0 – 1.0)', fontweight='bold')\n", "axes2[0].set_xlabel('')\n", "\n", "for i, v in enumerate(score_vals):\n", " axes2[0].text(\n", " i, v + 0.03, f'{v:.2f}',\n", " ha='center', fontweight='bold', fontsize=12\n", " )\n", "\n", "# ── Plot 2B: Batch Score Summary Bar ─────────────────────────\n", "summary_metrics = ['Faithfulness\\n(Batch)', 'Relevance\\n(Batch)']\n", "summary_means = [batch_results['mean_faithfulness'], batch_results['mean_relevance']]\n", "summary_stds = [batch_results['std_faithfulness'], batch_results['std_relevance']]\n", "\n", "if 'mean_correctness' in batch_results:\n", " summary_metrics.append('Correctness\\n(Batch)')\n", " summary_means.append(batch_results['mean_correctness'])\n", " summary_stds.append(batch_results['std_correctness'])\n", "\n", "summary_colors = ['#2ecc71' if m >= 0.8 else '#e74c3c' for m in summary_means]\n", "x2 = range(len(summary_metrics))\n", "\n", "axes2[1].bar(\n", " x2, summary_means, yerr=summary_stds, capsize=6,\n", " color=summary_colors, alpha=0.88, width=0.5,\n", " error_kw={'elinewidth': 2, 'capthick': 2}\n", ")\n", "axes2[1].set_xticks(list(x2))\n", "axes2[1].set_xticklabels(summary_metrics, fontsize=10)\n", "axes2[1].set_ylim(0, 1.2)\n", "axes2[1].set_ylabel('Mean Score (0.0 – 1.0)', fontweight='bold')\n", "axes2[1].set_title(\n", " f'Batch Evaluation Summary (n={batch_results[\"n\"]})',\n", " fontsize=12, pad=12\n", ")\n", "axes2[1].axhline(\n", " y=0.8, color='gray', linestyle='--',\n", " linewidth=1.2, alpha=0.7, label='Pass threshold (0.80)'\n", ")\n", "axes2[1].legend(fontsize=9)\n", "\n", "for i, (m, s) in enumerate(zip(summary_means, summary_stds)):\n", " axes2[1].text(\n", " i, m + s + 0.04, f'{m:.3f}',\n", " ha='center', fontweight='bold', fontsize=11\n", " )\n", "\n", "plt.tight_layout()\n", "telemetry_path = f'{VISUALS_DIR}/telemetry_dashboard.png'\n", "plt.savefig(telemetry_path, dpi=300, bbox_inches='tight')\n", "print(f'Telemetry dashboard saved: {telemetry_path}')\n", "plt.show()\n", "\n", "# ── Final Summary Print ───────────────────────────────────────\n", "print('\\n' + '='*65)\n", "print(' FINANCIAL INTELLIGENCE ENGINE — FINAL METRICS SUMMARY')\n", "print('='*65)\n", "print(f' Faithfulness (n={batch_results[\"n\"]}) : '\n", " f'{batch_results[\"mean_faithfulness\"]:.3f} ± {batch_results[\"std_faithfulness\"]:.3f} '\n", " f'| Pass rate: {batch_results[\"faithfulness_pass_rate\"]*100:.1f}%')\n", "print(f' Relevance (n={batch_results[\"n\"]}) : '\n", " f'{batch_results[\"mean_relevance\"]:.3f} ± {batch_results[\"std_relevance\"]:.3f}')\n", "if 'mean_correctness' in batch_results:\n", " print(f' Correctness (n={batch_results[\"n\"]}) : '\n", " f'{batch_results[\"mean_correctness\"]:.3f} ± {batch_results[\"std_correctness\"]:.3f}')\n", "print(f' Generator model : {agent.llm.model_name}')\n", "print(f' Evaluator model : {batch_results[\"evaluator_model\"]}')\n", "print(f' Retrieval method : Hybrid RRF (Dense ChromaDB + Sparse BM25)')\n", "print(f' Company balance : {dict(Counter(d.metadata.get(\"company\") for d in used_docs))}')\n", "print('='*65)\n", "print('All artifacts saved to Google Drive /artifacts/ directory.')" ] }, { "cell_type": "code", "execution_count": null, "id": "0a119ace", "metadata": {}, "outputs": [], "source": [] } ], "metadata": { "kernelspec": { "display_name": "Python 3 (ipykernel)", "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": "3.12.13" } }, "nbformat": 4, "nbformat_minor": 5 }