Merge pull request #876 from muthash/stephen/week5-exercise
[Bootcamp] Week 5 RAG Exercise (Stephen)
This commit is contained in:
Ed Donner
GitHub
@@ -0,0 +1,243 @@
|
|||||||
|
{
|
||||||
|
"cells": [
|
||||||
|
{
|
||||||
|
"cell_type": "markdown",
|
||||||
|
"id": "dfe37963-1af6-44fc-a841-8e462443f5e6",
|
||||||
|
"metadata": {},
|
||||||
|
"source": [
|
||||||
|
"## Expert Knowledge Worker\n",
|
||||||
|
"\n",
|
||||||
|
"### A question answering agent that is an expert knowledge worker\n",
|
||||||
|
"### To be used by Anyone on their LinkedIn data\n",
|
||||||
|
"The easiest and fastest way to obtain a copy of your LinkedIn data is to initiate a data download from your Settings & Privacy page:\n",
|
||||||
|
"\n",
|
||||||
|
"1. Click the Me icon at the top of your LinkedIn homepage.\n",
|
||||||
|
"2. Select Settings & Privacy from the dropdown.\n",
|
||||||
|
"3. Click the Data Privacy on the left rail.\n",
|
||||||
|
"4 .Under the How LinkedIn uses your data section, click Get a copy of your data.\n",
|
||||||
|
"5. Select the data that you’re looking for and Request archive.\n",
|
||||||
|
"\n",
|
||||||
|
"This project will use RAG (Retrieval Augmented Generation) to ensure our question/answering assistant has high accuracy."
|
||||||
|
]
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"cell_type": "code",
|
||||||
|
"execution_count": 17,
|
||||||
|
"id": "ba2779af-84ef-4227-9e9e-6eaf0df87e77",
|
||||||
|
"metadata": {},
|
||||||
|
"outputs": [],
|
||||||
|
"source": [
|
||||||
|
"# imports\n",
|
||||||
|
"\n",
|
||||||
|
"import os\n",
|
||||||
|
"import glob\n",
|
||||||
|
"from dotenv import load_dotenv\n",
|
||||||
|
"import gradio as gr\n",
|
||||||
|
"\n",
|
||||||
|
"from langchain.document_loaders import DirectoryLoader, TextLoader\n",
|
||||||
|
"from langchain.text_splitter import CharacterTextSplitter\n",
|
||||||
|
"from langchain.schema import Document\n",
|
||||||
|
"from langchain_openai import OpenAIEmbeddings, ChatOpenAI\n",
|
||||||
|
"from langchain_chroma import Chroma\n",
|
||||||
|
"import plotly.graph_objects as go\n",
|
||||||
|
"from langchain.memory import ConversationBufferMemory\n",
|
||||||
|
"from langchain.chains import ConversationalRetrievalChain\n",
|
||||||
|
"from langchain.embeddings import HuggingFaceEmbeddings\n",
|
||||||
|
"\n",
|
||||||
|
"import matplotlib.pyplot as plt\n",
|
||||||
|
"from sklearn.manifold import TSNE\n",
|
||||||
|
"import numpy as np\n",
|
||||||
|
"\n",
|
||||||
|
"MODEL = \"gpt-4o-mini\"\n",
|
||||||
|
"db_name = \"linkedin_db\"\n",
|
||||||
|
"\n",
|
||||||
|
"load_dotenv(override=True)\n",
|
||||||
|
"os.environ['OPENAI_API_KEY'] = os.getenv('OPENAI_API_KEY', 'your-key-if-not-using-env')"
|
||||||
|
]
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"cell_type": "code",
|
||||||
|
"execution_count": null,
|
||||||
|
"id": "730711a9-6ffe-4eee-8f48-d6cfb7314905",
|
||||||
|
"metadata": {},
|
||||||
|
"outputs": [],
|
||||||
|
"source": [
|
||||||
|
"# Read in documents using LangChain's loaders\n",
|
||||||
|
"# Put the chunks of data into a Vector Store (Chroma) that associates a Vector Embedding with each chunk\n",
|
||||||
|
"\n",
|
||||||
|
"folders = glob.glob(\"linkedin-base/*\")\n",
|
||||||
|
"\n",
|
||||||
|
"def add_metadata(doc, doc_type):\n",
|
||||||
|
" doc.metadata[\"doc_type\"] = doc_type\n",
|
||||||
|
" return doc\n",
|
||||||
|
"\n",
|
||||||
|
"text_loader_kwargs = {'encoding': 'utf-8'}\n",
|
||||||
|
"\n",
|
||||||
|
"documents = []\n",
|
||||||
|
"for folder in folders:\n",
|
||||||
|
" doc_type = os.path.basename(folder)\n",
|
||||||
|
" loader = DirectoryLoader(folder, glob=\"**/*.md\", loader_cls=TextLoader, loader_kwargs=text_loader_kwargs)\n",
|
||||||
|
" folder_docs = loader.load()\n",
|
||||||
|
" documents.extend([add_metadata(doc, doc_type) for doc in folder_docs])\n",
|
||||||
|
"\n",
|
||||||
|
"text_splitter = CharacterTextSplitter(chunk_size=500, chunk_overlap=100)\n",
|
||||||
|
"chunks = text_splitter.split_documents(documents)\n",
|
||||||
|
"\n",
|
||||||
|
"embeddings = OpenAIEmbeddings()\n",
|
||||||
|
"\n",
|
||||||
|
"if os.path.exists(db_name):\n",
|
||||||
|
" Chroma(persist_directory=db_name, embedding_function=embeddings).delete_collection()\n",
|
||||||
|
"\n",
|
||||||
|
"vectorstore = Chroma.from_documents(documents=chunks, embedding=embeddings, persist_directory=db_name)\n",
|
||||||
|
"\n",
|
||||||
|
"collection = vectorstore._collection\n",
|
||||||
|
"count = collection.count()\n",
|
||||||
|
"\n",
|
||||||
|
"sample_embedding = collection.get(limit=1, include=[\"embeddings\"])[\"embeddings\"][0]\n",
|
||||||
|
"dimensions = len(sample_embedding)\n",
|
||||||
|
"\n",
|
||||||
|
"\n",
|
||||||
|
"print(f\"Total number of chunks: {len(chunks)}\")\n",
|
||||||
|
"print(f\"Document types found: {set(doc.metadata['doc_type'] for doc in documents)}\")\n",
|
||||||
|
"print(f\"Vectorstore created with {vectorstore._collection.count()} documents\")\n",
|
||||||
|
"print(f\"There are {count:,} vectors with {dimensions:,} dimensions in the vector store\")"
|
||||||
|
]
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"cell_type": "code",
|
||||||
|
"execution_count": null,
|
||||||
|
"id": "b98adf5e-d464-4bd2-9bdf-bc5b6770263b",
|
||||||
|
"metadata": {},
|
||||||
|
"outputs": [],
|
||||||
|
"source": [
|
||||||
|
"# 2D scatter plot\n",
|
||||||
|
"\n",
|
||||||
|
"result = collection.get(include=['embeddings', 'documents', 'metadatas'])\n",
|
||||||
|
"vectors = np.array(result['embeddings'])\n",
|
||||||
|
"documents = result['documents']\n",
|
||||||
|
"metadatas = result['metadatas']\n",
|
||||||
|
"doc_types = [metadata['doc_type'] for metadata in metadatas]\n",
|
||||||
|
"colors = [['blue', 'green', 'red'][['connections', 'recommendations', 'profiles'].index(t)] for t in doc_types]\n",
|
||||||
|
"\n",
|
||||||
|
"n = vectors.shape[0]\n",
|
||||||
|
"if n < 3:\n",
|
||||||
|
" raise ValueError(f\"t-SNE needs at least 3 samples, got {n}\")\n",
|
||||||
|
"\n",
|
||||||
|
"perp = max(5.0, min(30.0, (n - 1) / 3.0)) # always < n, within [5, 30]\n",
|
||||||
|
"\n",
|
||||||
|
"tsne = TSNE(n_components=2, random_state=42, perplexity=perp)\n",
|
||||||
|
"reduced_vectors = tsne.fit_transform(vectors)\n",
|
||||||
|
"\n",
|
||||||
|
"fig = go.Figure(data=[go.Scatter(\n",
|
||||||
|
" x=reduced_vectors[:, 0],\n",
|
||||||
|
" y=reduced_vectors[:, 1],\n",
|
||||||
|
" mode='markers',\n",
|
||||||
|
" marker=dict(size=5, color=colors, opacity=0.8),\n",
|
||||||
|
" text=[f\"Type: {t}<br>Text: {d[:100]}...\" for t, d in zip(doc_types, documents)],\n",
|
||||||
|
" hoverinfo='text'\n",
|
||||||
|
")])\n",
|
||||||
|
"\n",
|
||||||
|
"fig.update_layout(\n",
|
||||||
|
" title='2D Chroma Vector Store Visualization',\n",
|
||||||
|
" scene=dict(xaxis_title='x',yaxis_title='y'),\n",
|
||||||
|
" width=800,\n",
|
||||||
|
" height=600,\n",
|
||||||
|
" margin=dict(r=20, b=10, l=10, t=40)\n",
|
||||||
|
")\n",
|
||||||
|
"\n",
|
||||||
|
"fig.show()"
|
||||||
|
]
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"cell_type": "code",
|
||||||
|
"execution_count": null,
|
||||||
|
"id": "e1418e88-acd5-460a-bf2b-4e6efc88e3dd",
|
||||||
|
"metadata": {},
|
||||||
|
"outputs": [],
|
||||||
|
"source": [
|
||||||
|
"# 3D scatter plot!\n",
|
||||||
|
"\n",
|
||||||
|
"n = vectors.shape[0]\n",
|
||||||
|
"if n < 3:\n",
|
||||||
|
" raise ValueError(f\"t-SNE needs at least 3 samples, got {n}\")\n",
|
||||||
|
"\n",
|
||||||
|
"perp = max(5.0, min(30.0, (n - 1) / 3.0))\n",
|
||||||
|
"\n",
|
||||||
|
"tsne = TSNE(n_components=3, random_state=42, perplexity=perp)\n",
|
||||||
|
"reduced_vectors = tsne.fit_transform(vectors)\n",
|
||||||
|
"\n",
|
||||||
|
"fig = go.Figure(data=[go.Scatter3d(\n",
|
||||||
|
" x=reduced_vectors[:, 0],\n",
|
||||||
|
" y=reduced_vectors[:, 1],\n",
|
||||||
|
" z=reduced_vectors[:, 2],\n",
|
||||||
|
" mode='markers',\n",
|
||||||
|
" marker=dict(size=5, color=colors, opacity=0.8),\n",
|
||||||
|
" text=[f\"Type: {t}<br>Text: {d[:100]}...\" for t, d in zip(doc_types, documents)],\n",
|
||||||
|
" hoverinfo='text'\n",
|
||||||
|
")])\n",
|
||||||
|
"\n",
|
||||||
|
"fig.update_layout(\n",
|
||||||
|
" title='3D Chroma Vector Store Visualization',\n",
|
||||||
|
" scene=dict(xaxis_title='x', yaxis_title='y', zaxis_title='z'),\n",
|
||||||
|
" width=900,\n",
|
||||||
|
" height=700,\n",
|
||||||
|
" margin=dict(r=20, b=10, l=10, t=40)\n",
|
||||||
|
")\n",
|
||||||
|
"\n",
|
||||||
|
"fig.show()"
|
||||||
|
]
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"cell_type": "code",
|
||||||
|
"execution_count": null,
|
||||||
|
"id": "2136153b-d2f6-4c58-a0e3-78c3a932cf55",
|
||||||
|
"metadata": {},
|
||||||
|
"outputs": [],
|
||||||
|
"source": [
|
||||||
|
"# The main Langchain Abstraction are: Memory, LLM, and Retriever\n",
|
||||||
|
"llm = ChatOpenAI(temperature=0.7, model_name=MODEL)\n",
|
||||||
|
"\n",
|
||||||
|
"memory = ConversationBufferMemory(memory_key='chat_history', return_messages=True)\n",
|
||||||
|
"retriever = vectorstore.as_retriever(search_kwargs={\"k\": 25})\n",
|
||||||
|
"conversation_chain = ConversationalRetrievalChain.from_llm(llm=llm, retriever=retriever, memory=memory)\n",
|
||||||
|
"\n",
|
||||||
|
"def chat(question, history):\n",
|
||||||
|
" result = conversation_chain.invoke({\"question\": question})\n",
|
||||||
|
" return result[\"answer\"]\n",
|
||||||
|
"\n",
|
||||||
|
"with gr.Blocks(theme=\"gradio/monochrome\") as ui:\n",
|
||||||
|
" gr.Markdown(\n",
|
||||||
|
" \"\"\"\n",
|
||||||
|
" <h2 style=\"color: #f5f5f5;\">Linkedin Knowledge Worker</h2>\n",
|
||||||
|
" <p style=\"color: #f5f5f5;\">Chat with your auto-generated Linkedin knowledge base </p>\n",
|
||||||
|
" \"\"\",\n",
|
||||||
|
" elem_id=\"title\"\n",
|
||||||
|
" )\n",
|
||||||
|
" gr.ChatInterface(chat, type=\"messages\")\n",
|
||||||
|
"\n",
|
||||||
|
"ui.launch(inbrowser=True)"
|
||||||
|
]
|
||||||
|
}
|
||||||
|
],
|
||||||
|
"metadata": {
|
||||||
|
"kernelspec": {
|
||||||
|
"display_name": ".venv",
|
||||||
|
"language": "python",
|
||||||
|
"name": "python3"
|
||||||
|
},
|
||||||
|
"language_info": {
|
||||||
|
"codemirror_mode": {
|
||||||
|
"name": "ipython",
|
||||||
|
"version": 3
|
||||||
|
},
|
||||||
|
"file_extension": ".py",
|
||||||
|
"mimetype": "text/x-python",
|
||||||
|
"name": "python",
|
||||||
|
"nbconvert_exporter": "python",
|
||||||
|
"pygments_lexer": "ipython3",
|
||||||
|
"version": "3.12.12"
|
||||||
|
}
|
||||||
|
},
|
||||||
|
"nbformat": 4,
|
||||||
|
"nbformat_minor": 5
|
||||||
|
}
|
||||||
Reference in New Issue
Block a user