Spaces:

Ekimetrics
/

climate-question-answering

Running

File size: 146,496 Bytes

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The autoreload extension is already loaded. To reload it, use:\n",
      "  %reload_ext autoreload\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd \n",
    "import numpy as np\n",
    "import os\n",
    "\n",
    "%load_ext autoreload\n",
    "%autoreload 2\n",
    "\n",
    "import sys\n",
    "sys.path.append(os.path.dirname(os.getcwd()))\n",
    "\n",
    "from dotenv import load_dotenv\n",
    "load_dotenv()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "from climateqa.engine.llm import get_llm\n",
    "from climateqa.engine.vectorstore import get_pinecone_vectorstore\n",
    "from climateqa.engine.embeddings import get_embeddings_function\n",
    "from climateqa.engine.reranker import get_reranker\n",
    "from climateqa.engine.graph import make_graph_agent, display_graph\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## LLM"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "AIMessage(content='Hello! How can I assist you today?', response_metadata={'finish_reason': 'stop'}, id='run-d2d534d6-0560-4819-9aac-ecf073cb7b7a-0')"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from climateqa.engine.llm import get_llm\n",
    "\n",
    "llm = get_llm(provider=\"openai\")\n",
    "llm.invoke(\"Say Hello !\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Retriever "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from climateqa.engine.vectorstore import get_pinecone_vectorstore\n",
    "from climateqa.engine.embeddings import get_embeddings_function\n",
    "\n",
    "question = \"What is the impact of climate change on the environment?\"\n",
    "\n",
    "embeddings_function = get_embeddings_function()\n",
    "vectorstore_ipcc = get_pinecone_vectorstore(embeddings_function)\n",
    "docs_question = vectorstore_ipcc.search(query = question, search_type=\"similarity\")\n",
    "docs_question"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# optional filters\n",
    "sources_owid = [\"OWID\"]\n",
    "filters = {}\n",
    "filters[\"source\"] = {\"$in\": sources_owid}\n",
    "\n",
    "# vectorestore_graphs\n",
    "vectorstore_graphs = get_pinecone_vectorstore(embeddings_function, index_name = os.getenv(\"PINECONE_API_INDEX_OWID\"), text_key=\"title\")\n",
    "owid_graphs = vectorstore_graphs.search(query = question, search_type=\"similarity\")\n",
    "owid_graphs = vectorstore_graphs.similarity_search_with_score(query = question, filter=filters, k=5)\n",
    "owid_graphs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "vectorstore_region = get_pinecone_vectorstore(embeddings_function, index_name=os.getenv(\"PINECONE_API_INDEX_REGION\"))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Reranker"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from climateqa.engine.reranker import get_reranker\n",
    "from climateqa.engine.reranker import rerank_docs\n",
    "\n",
    "reranker = get_reranker(\"nano\")\n",
    "reranked_docs_question = rerank_docs(reranker,docs_question,question)\n",
    "reranked_docs_question"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Graph"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loading embeddings model:  BAAI/bge-base-en-v1.5\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/tim/ai4s/climate_qa/climate-question-answering/climateqa/engine/vectorstore.py:34: LangChainDeprecationWarning: The class `Pinecone` was deprecated in LangChain 0.0.18 and will be removed in 0.3.0. An updated version of the class exists in the langchain-pinecone package and should be used instead. To use it run `pip install -U langchain-pinecone` and import as `from langchain_pinecone import Pinecone`.\n",
      "  vectorstore = PineconeVectorstore(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loading FlashRankRanker model ms-marco-TinyBERT-L-2-v2\n",
      "Loading model FlashRank model ms-marco-TinyBERT-L-2-v2...\n"
     ]
    },
    {
     "data": {
      "image/jpeg": "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",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from climateqa.engine.graph import make_graph_agent, display_graph\n",
    "\n",
    "llm = get_llm(provider=\"openai\")\n",
    "embeddings_function = get_embeddings_function()\n",
    "vectorstore_ipcc = get_pinecone_vectorstore(embeddings_function)\n",
    "vectorstore_graphs = get_pinecone_vectorstore(embeddings_function, index_name = os.getenv(\"PINECONE_API_INDEX_OWID\"), text_key=\"title\")\n",
    "vectorstore_region = get_pinecone_vectorstore(embeddings_function, index_name=os.getenv(\"PINECONE_API_INDEX_REGION\"))\n",
    "reranker = get_reranker(\"nano\")\n",
    "\n",
    "app = make_graph_agent(llm=llm, vectorstore_ipcc=vectorstore_ipcc, vectorstore_graphs=vectorstore_graphs, vectorstore_region=vectorstore_region, reranker=reranker)\n",
    "display_graph(app)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "from climateqa.engine.graph import search \n",
    "\n",
    "from climateqa.engine.chains.intent_categorization import make_intent_categorization_node\n",
    "\n",
    "\n",
    "from climateqa.engine.chains.answer_chitchat import make_chitchat_node\n",
    "from climateqa.engine.chains.answer_ai_impact import make_ai_impact_node\n",
    "from climateqa.engine.chains.query_transformation import make_query_transform_node\n",
    "from climateqa.engine.chains.translation import make_translation_node\n",
    "from climateqa.engine.chains.retrieve_documents import make_IPx_retriever_node, make_POC_retriever_node\n",
    "from climateqa.engine.chains.answer_rag import make_rag_node\n",
    "from climateqa.engine.chains.graph_retriever import make_graph_retriever_node\n",
    "from climateqa.engine.chains.chitchat_categorization import make_chitchat_intent_categorization_node\n",
    "from climateqa.engine.chains.prompts import audience_prompts\n",
    "from climateqa.engine.graph import route_intent\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "inial_state = {\n",
    "    # \"user_input\": \"What is the impact of climate change on the environment?\", \n",
    "    \"user_input\": \"Quel est l'impact du changement climatique sur Bordeaux ?\",\n",
    "    \"audience\" : audience_prompts[\"general\"],\n",
    "    # \"sources_input\":[\"IPCC\"],\n",
    "    \"relevant_content_sources_selection\": [\"Figures (IPCC/IPBES)\",\"POC region\"],\n",
    "    \"search_only\" : False,\n",
    "    \"reports\": [],\n",
    "}\n",
    "state=inial_state.copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "cat_node = make_intent_categorization_node(llm)\n",
    "state.update(cat_node(inial_state))\n",
    "state"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [],
   "source": [
    "# state.update(search(state))\n",
    "# state"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {},
   "outputs": [],
   "source": [
    "intent = route_intent(state)\n",
    "\n",
    "if route_intent(state) == \"translate_query\":\n",
    "    make_translation_node(llm)(state)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "state.update(make_query_transform_node(llm)(state))\n",
    "state"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "state.update(make_graph_retriever_node(vectorstore_graphs, reranker)(state))\n",
    "state"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "retriever_node = make_POC_retriever_node(vectorstore_ipcc, reranker, llm)\n",
    "retriever_node"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "new_state"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "new_state = state.copy()\n",
    "evolutions_states = []\n",
    "while len(new_state[\"questions_list\"])>0:    \n",
    "    async for temp_state in retriever_node.astream(new_state):\n",
    "        evolutions_states.append(temp_state)\n",
    "        new_state.update(temp_state)\n",
    "        print(temp_state)\n",
    "# new_state"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "answer_rag = await make_rag_node(llm)(new_state,{})\n",
    "new_state.update(answer_rag)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# stream event of the whole chain"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:pinecone_plugin_interface.logging:Discovering subpackages in _NamespacePath(['/home/tim/anaconda3/envs/climateqa/lib/python3.11/site-packages/pinecone_plugins'])\n",
      "INFO:pinecone_plugin_interface.logging:Looking for plugins in pinecone_plugins.inference\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:pinecone_plugin_interface.logging:Installing plugin inference into Pinecone\n",
      "INFO:sentence_transformers.SentenceTransformer:Load pretrained SentenceTransformer: BAAI/bge-base-en-v1.5\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loading embeddings model:  BAAI/bge-base-en-v1.5\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:sentence_transformers.SentenceTransformer:Use pytorch device_name: cpu\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "{'user_input': 'What will be the precipitation in Bordeaux in 2050?',\n",
       " 'audience': 'the general public who know the basics in science and climate change and want to learn more about it without technical terms. Still use references to passages.',\n",
       " 'sources_input': ['IPCC'],\n",
       " 'relevant_content_sources_selection': [],\n",
       " 'search_only': False,\n",
       " 'reports': []}"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "from climateqa.engine.graph import make_graph_agent, display_graph\n",
    "from climateqa.engine.chains.prompts import audience_prompts\n",
    "\n",
    "\n",
    "inial_state = {\n",
    "    # \"user_input\": \"What is the impact of climate change on the environment?\", \n",
    "    # \"user_input\": \"What is the impact of climate  in Bordeaux\", \n",
    "    \"user_input\": \"What will be the precipitation in Bordeaux in 2050?\", \n",
    "    \"audience\" : audience_prompts[\"general\"],\n",
    "    \"sources_input\":[\"IPCC\"],\n",
    "    # \"relevant_content_sources_selection\": [\"Figures (IPCC/IPBES)\",\"POC region\"],\n",
    "    \"relevant_content_sources_selection\": [],\n",
    "    \"search_only\" : False,\n",
    "    \"reports\": [],\n",
    "}\n",
    "app = make_graph_agent(llm=llm, vectorstore_ipcc=vectorstore_ipcc, vectorstore_graphs=vectorstore_graphs, vectorstore_region=vectorstore_region, reranker=reranker)\n",
    "\n",
    "inial_state"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "---- Categorize_message ----\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:httpx:HTTP Request: POST https://api.openai.com/v1/chat/completions \"HTTP/1.1 200 OK\"\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "\n",
      "Output intent categorization: {'intent': 'search'}\n",
      "\n",
      "---- Transform query ----\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:httpx:HTTP Request: POST https://api.openai.com/v1/chat/completions \"HTTP/1.1 200 OK\"\n",
      "INFO:httpx:HTTP Request: POST https://api.openai.com/v1/chat/completions \"HTTP/1.1 200 OK\"\n",
      "INFO:httpx:HTTP Request: POST https://api.openai.com/v1/chat/completions \"HTTP/1.1 200 OK\"\n",
      "INFO:httpx:HTTP Request: POST https://api.openai.com/v1/chat/completions \"HTTP/1.1 200 OK\"\n",
      "INFO:httpx:HTTP Request: POST https://api.openai.com/v1/chat/completions \"HTTP/1.1 200 OK\"\n",
      "INFO:httpx:HTTP Request: POST https://api.openai.com/v1/chat/completions \"HTTP/1.1 200 OK\"\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "---- Retrieving data from DRIAS ----\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:httpx:HTTP Request: POST https://api.openai.com/v1/chat/completions \"HTTP/1.1 200 OK\"\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SQL Prompt: [{'role': 'system', 'content': \"You are a SQLite expert. Please help to generate a SQL query to answer the question. Your response should ONLY be based on the given context and follow the response guidelines and format instructions. \\n===Tables \\n\\n            CREATE TABLE Winter_precipitation_total (\\n                year INT, \\n month INT, \\n  day INT \\n,\\n                x FLOAT,\\n                y FLOAT,\\n                LambertParisII VARCHAR(255),\\n                RR FLOAT,\\n                lat FLOAT,\\n                lon FLOAT,\\n            );\\n        \\n\\n\\n                CREATE TABLE Summer_precipitation_total (\\n                    year INT, \\n month INT, \\n  day INT \\n,\\n                    x FLOAT,\\n                    y FLOAT,\\n                    LambertParisII VARCHAR(255),\\n                    RR FLOAT,\\n                    lat FLOAT,\\n                    lon FLOAT,\\n                );\\n            \\n\\n\\n===Additional Context \\n\\n\\n                This table contains information on the intensity of extreme precipitation in the past and the future,\\n                which represents the maximum value of total annual precipitation.\\n                The variables are as follows:\\n                - 'y' and 'x': Lambert Paris II coordinates for the location.\\n                - year: Year of the observation.\\n\\n              - month : Month of the observation.\\n\\n              - day: Day of the observation.\\n\\n                - 'LambertParisII': Indicates that the x, y coordinates are in the Lambert Paris II projection.\\n                - 'lat' and 'lon': Latitude and longitude of the location.\\n                - 'RX1d': Intensity of extreme precipitation (maximum annual total precipitation).\\n                \\n\\n\\n            This table contains the cumulative winter precipitation in the past and the future.\\n            The columns include:\\n            - year: Year of the observation.\\n\\n              - month : Month of the observation.\\n\\n              - day: Day of the observation.\\n\\n            - `x`: Coordinate in the Lambert II projection for the location.\\n            - `y`: Coordinate in the Lambert II projection for the location.\\n            - `LambertParisII`: Indicates that the x, y coordinates are in the Lambert Paris II projection.\\n            - `RR`: Cumulative winter precipitation.\\n            - `lat`: Geographic latitude of the location.\\n            - `lon`: Geographic longitude of the location.\\n            \\n\\n===Response Guidelines \\n1. If the provided context is sufficient, please generate a valid SQL query without any explanations for the question. \\n2. If the provided context is almost sufficient but requires knowledge of a specific string in a particular column, please generate an intermediate SQL query to find the distinct strings in that column. Prepend the query with a comment saying intermediate_sql \\n3. If the provided context is insufficient, please explain why it can't be generated. \\n4. Please use the most relevant table(s). \\n5. If the question has been asked and answered before, please repeat the answer exactly as it was given before. \\n6. Ensure that the output SQL is SQLite-compliant and executable, and free of syntax errors. \\n\"}, {'role': 'user', 'content': 'What will be the precipitation in lat, long : (44.841225, -0.5800364) in 2050?'}]\n",
      "Using model gpt-4o-mini for 802.25 tokens (approx)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:httpx:HTTP Request: POST https://api.openai.com/v1/chat/completions \"HTTP/1.1 200 OK\"\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "LLM Response: intermediate_sql\n",
      "```sql\n",
      "SELECT DISTINCT year FROM Winter_precipitation_total WHERE lat = 44.841225 AND lon = -0.5800364;\n",
      "```\n",
      "Extracted SQL: SELECT DISTINCT year FROM Winter_precipitation_total WHERE lat = 44.841225 AND lon = -0.5800364;\n",
      "Running Intermediate SQL: SELECT DISTINCT year FROM Winter_precipitation_total WHERE lat = 44.841225 AND lon = -0.5800364;\n",
      "Final SQL Prompt: [{'role': 'system', 'content': \"You are a SQLite expert. Please help to generate a SQL query to answer the question. Your response should ONLY be based on the given context and follow the response guidelines and format instructions. \\n===Tables \\n\\n            CREATE TABLE Winter_precipitation_total (\\n                year INT, \\n month INT, \\n  day INT \\n,\\n                x FLOAT,\\n                y FLOAT,\\n                LambertParisII VARCHAR(255),\\n                RR FLOAT,\\n                lat FLOAT,\\n                lon FLOAT,\\n            );\\n        \\n\\n\\n                CREATE TABLE Summer_precipitation_total (\\n                    year INT, \\n month INT, \\n  day INT \\n,\\n                    x FLOAT,\\n                    y FLOAT,\\n                    LambertParisII VARCHAR(255),\\n                    RR FLOAT,\\n                    lat FLOAT,\\n                    lon FLOAT,\\n                );\\n            \\n\\n\\n===Additional Context \\n\\n\\n                This table contains information on the intensity of extreme precipitation in the past and the future,\\n                which represents the maximum value of total annual precipitation.\\n                The variables are as follows:\\n                - 'y' and 'x': Lambert Paris II coordinates for the location.\\n                - year: Year of the observation.\\n\\n              - month : Month of the observation.\\n\\n              - day: Day of the observation.\\n\\n                - 'LambertParisII': Indicates that the x, y coordinates are in the Lambert Paris II projection.\\n                - 'lat' and 'lon': Latitude and longitude of the location.\\n                - 'RX1d': Intensity of extreme precipitation (maximum annual total precipitation).\\n                \\n\\n\\n            This table contains the cumulative winter precipitation in the past and the future.\\n            The columns include:\\n            - year: Year of the observation.\\n\\n              - month : Month of the observation.\\n\\n              - day: Day of the observation.\\n\\n            - `x`: Coordinate in the Lambert II projection for the location.\\n            - `y`: Coordinate in the Lambert II projection for the location.\\n            - `LambertParisII`: Indicates that the x, y coordinates are in the Lambert Paris II projection.\\n            - `RR`: Cumulative winter precipitation.\\n            - `lat`: Geographic latitude of the location.\\n            - `lon`: Geographic longitude of the location.\\n            \\n\\nThe following is a pandas DataFrame with the results of the intermediate SQL query SELECT DISTINCT year FROM Winter_precipitation_total WHERE lat = 44.841225 AND lon = -0.5800364;: \\n| year   |\\n|--------|\\n\\n===Response Guidelines \\n1. If the provided context is sufficient, please generate a valid SQL query without any explanations for the question. \\n2. If the provided context is almost sufficient but requires knowledge of a specific string in a particular column, please generate an intermediate SQL query to find the distinct strings in that column. Prepend the query with a comment saying intermediate_sql \\n3. If the provided context is insufficient, please explain why it can't be generated. \\n4. Please use the most relevant table(s). \\n5. If the question has been asked and answered before, please repeat the answer exactly as it was given before. \\n6. Ensure that the output SQL is SQLite-compliant and executable, and free of syntax errors. \\n\"}, {'role': 'user', 'content': 'What will be the precipitation in lat, long : (44.841225, -0.5800364) in 2050?'}]\n",
      "Using model gpt-4o-mini for 853.5 tokens (approx)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:httpx:HTTP Request: POST https://api.openai.com/v1/chat/completions \"HTTP/1.1 200 OK\"\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "LLM Response: ```sql\n",
      "SELECT RR FROM Winter_precipitation_total WHERE lat = 44.841225 AND lon = -0.5800364 AND year = 2050;\n",
      "```\n",
      "Extracted SQL: SELECT RR FROM Winter_precipitation_total WHERE lat = 44.841225 AND lon = -0.5800364 AND year = 2050;\n",
      "Using model gpt-4o-mini for 188.0 tokens (approx)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:httpx:HTTP Request: POST https://api.openai.com/v1/chat/completions \"HTTP/1.1 200 OK\"\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[(300.7206382352941,), (300.7206382352941,)]\n"
     ]
    }
   ],
   "source": [
    "event_list = app.astream_events(inial_state, version = \"v1\")\n",
    "static_event_list = []\n",
    "async for event in event_list:\n",
    "    static_event_list.append(event)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "---- Categorize_message ----\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:httpx:HTTP Request: POST https://api.openai.com/v1/chat/completions \"HTTP/1.1 200 OK\"\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "\n",
      "Output intent categorization: {'intent': 'search'}\n",
      "\n",
      "---- Transform query ----\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:httpx:HTTP Request: POST https://api.openai.com/v1/chat/completions \"HTTP/1.1 200 OK\"\n",
      "INFO:httpx:HTTP Request: POST https://api.openai.com/v1/chat/completions \"HTTP/1.1 200 OK\"\n",
      "INFO:httpx:HTTP Request: POST https://api.openai.com/v1/chat/completions \"HTTP/1.1 200 OK\"\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Here range(0, 4) []\n",
      "Retrieve documents for question: What are the effects of climate change on Bordeaux's environment?\n",
      "Here range(0, 4) []\n",
      "Retrieve documents for question: What are the effects of climate change on Bordeaux's environment?\n",
      "---- Retrieve documents from IPx----\n",
      "Updated state with question  0  added  2  documents\n",
      "---- Retrieve documents from POC----\n",
      "Updated state with question  1  added  3  documents\n",
      "Here range(0, 4) [1, 0]\n",
      "Retrieve documents for question: How is climate change affecting agriculture and wine production in Bordeaux?\n",
      "Here range(0, 4) [1, 0]\n",
      "Retrieve documents for question: How is climate change affecting agriculture and wine production in Bordeaux?\n",
      "---- Retrieve documents from POC----\n",
      "Updated state with question  3  added  3  documents\n",
      "---- Retrieve documents from IPx----\n",
      "Updated state with question  2  added  2  documents\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:langchain_core.callbacks.manager:Error in LogStreamCallbackHandler.on_chain_start callback: ValidationError(model='Run', errors=[{'loc': ('__root__',), 'msg': \"argument of type 'NoneType' is not iterable\", 'type': 'type_error'}])\n",
      "WARNING:langchain_core.callbacks.manager:Error in LangChainTracer.on_chain_start callback: ValidationError(model='Run', errors=[{'loc': ('__root__',), 'msg': \"argument of type 'NoneType' is not iterable\", 'type': 'type_error'}])\n",
      "WARNING:langchain_core.callbacks.manager:Error in LogStreamCallbackHandler.on_chain_end callback: TracerException('No indexed run ID ec20638e-5216-47fa-ae26-26627bae4c0d.')\n",
      "WARNING:langchain_core.callbacks.manager:Error in LangChainTracer.on_chain_end callback: TracerException('No indexed run ID ec20638e-5216-47fa-ae26-26627bae4c0d.')\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Route :  ['answer_rag']\n",
      "---- Answer RAG ----\n",
      "Sources used : Acclimatera CC NA - page 297\n",
      "Acclimatera CC NA - page 458\n",
      "Acclimatera CC NA - page 459\n",
      "IPCC SR CCL SPM - page 16\n",
      "IPCC AR6 WGII FR - page 754\n",
      "Acclimatera CC NA - page 459\n",
      "Acclimatera CC NA - page 279\n",
      "Acclimatera CC NA - page 458\n",
      "IPCC SR CCL SPM - page 10\n",
      "IPCC AR6 WGII FR - page 1856\n",
      "CONTEXT LENGTH :  7737\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:httpx:HTTP Request: POST https://api.openai.com/v1/chat/completions \"HTTP/1.1 200 OK\"\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RAG elapsed time:  17.733213186264038\n",
      "Answer size :  1936\n"
     ]
    }
   ],
   "source": [
    "# Get the answer at the end\n",
    "from climateqa.handle_stream_events import stream_answer\n",
    "event_list = app.astream_events(inial_state, version = \"v1\")\n",
    "history = []\n",
    "start_streaming = False\n",
    "answer_message_content = \"\"\n",
    "async for event in event_list:\n",
    "\n",
    "    if \"langgraph_node\" in event[\"metadata\"]:\n",
    "        node = event[\"metadata\"][\"langgraph_node\"]\n",
    "\n",
    "        if (event[\"name\"] != \"transform_query\" and \n",
    "                      event[\"event\"] == \"on_chat_model_stream\" and\n",
    "                      node in [\"answer_rag\",\"answer_rag_no_docs\", \"answer_search\", \"answer_chitchat\"]):\n",
    "                    history, start_streaming, answer_message_content = stream_answer(\n",
    "                        history, event, start_streaming, answer_message_content\n",
    "                    )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Test events logs\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "inial_state = {'user_input': 'What is the impact of climate  in Bordeaux',\n",
    " 'audience': 'the general public who know the basics in science and climate change and want to learn more about it without technical terms. Still use references to passages.',\n",
    " 'sources_input': ['IPCC'],\n",
    " 'relevant_content_sources_selection': ['Figures (IPCC/IPBES)', 'POC region'],\n",
    " 'search_only': False,\n",
    " 'reports': []\n",
    " }"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Get the answer at the end\n",
    "from climateqa.handle_stream_events import stream_answer\n",
    "app = make_graph_agent(llm=llm, vectorstore_ipcc=vectorstore_ipcc, vectorstore_graphs=vectorstore_graphs, vectorstore_region=vectorstore_region, reranker=reranker)\n",
    "\n",
    "event_list = app.astream_events(inial_state, version = \"v1\")\n",
    "history = []\n",
    "start_streaming = False\n",
    "answer_message_content = \"\"\n",
    "static_event_list = []\n",
    "async for event in event_list:\n",
    "    static_event_list.append(event)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_static_events = pd.DataFrame(static_event_list)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_static_events.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_static_events[\"name\"].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "selected_events = df_static_events[\n",
    "    (df_static_events[\"event\"] == \"on_chain_end\") &\n",
    "    (df_static_events[\"name\"].isin([\"retrieve_documents\", \"retrieve_local_data\", \"retrieve_POC_docs_node\",\"retrieve_IPx_docs\"]))\n",
    "    # (df_static_events[\"data\"].apply(lambda x: x[\"output\"] is not None))\n",
    "]\n",
    "selected_events"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# selected_events[selected_events[\"data\"].apply(lambda x : \"output\" in x and x[\"output\"] is not None)]\n",
    "selected_events[\"data\"].apply(lambda x : x[\"output\"][\"documents\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "selected_events = df_static_events[\n",
    "    (df_static_events[\"event\"] == \"on_chain_end\") &\n",
    "    (df_static_events[\"name\"].isin([\"answer_search\"]))\n",
    "    # (df_static_events[\"data\"].apply(lambda x: x[\"output\"] is not None))\n",
    "]\n",
    "selected_events[\"metadata\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "selected_events[\"data\"].iloc[0][\"input\"][\"related_contents\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "selected_events[\"data\"].apply(lambda x : x[\"output\"]).iloc[2]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "selected_events.iloc[0][\"data\"].values()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "selected_events.iloc[1][\"data\"].values()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "list(selected_events.iloc[0][\"data\"].values())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "list(selected_events.iloc[1][\"data\"].values())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "list(selected_events.iloc[2][\"data\"].values())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "list(selected_events.iloc[3][\"data\"].values())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [],
   "source": [
    "# import json\n",
    "\n",
    "# print(json.dumps(list(selected_events.iloc[1][\"data\"].values()), indent=4))\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "data_values = selected_events.iloc[1][\"data\"].values()\n",
    "formatted_data = json.dumps(list(data_values)[0], indent=4)\n",
    "print(formatted_data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from pprint import pprint\n",
    "import json\n",
    "selected_events.iloc[2][\"data\"].values()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "selected_events.iloc[3][\"data\"].values()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_static_events[df_static_events[\"name\"] == \"retrieve_POC_docs_node\"].iloc[0]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "climateqa",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}