preprocess_chroma.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Install all the required libraries as needed.\n",
    "# !pip install langchain\n",
    "# !pip install unstructured\n",
    "# !pip install sentence_transformers\n",
    "# !pip install huggingface_hub\n",
    "# !pip install \"unstructured[pdf]\"\n",
    "# !pip install chromadb"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import necessary libraries.\n",
    "from langchain.document_loaders import DirectoryLoader\n",
    "from langchain.text_splitter import CharacterTextSplitter\n",
    "from langchain.vectorstores import Chroma\n",
    "from langchain.embeddings import HuggingFaceEmbeddings"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create directory loader object to load the data from the directory.\n",
    "pdfLoader = DirectoryLoader('./data/')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Initial the document loader to load the data from the directory.\n",
    "documents = []\n",
    "documents.extend(pdfLoader.load())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "You have 1 document(s) in your data folder.\n"
     ]
    }
   ],
   "source": [
    "# Display the number of documents in the directory.\n",
    "print(f'You have {len(documents)} document(s) in your data folder.')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Created a chunk of size 1164, which is longer than the specified 1000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "202\n"
     ]
    }
   ],
   "source": [
    "# Split the documents into chunks of 1000 characters with 200 characters overlap.\n",
    "text_splitter = CharacterTextSplitter(chunk_size=1000, chunk_overlap=200)\n",
    "documents = text_splitter.split_documents(documents)\n",
    "print(len(documents))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define to variables to use \"sentence-transformers/all-MiniLM-L6-v2\" embedding model from HuggingFace.\n",
    "# https://huggingface.co/blog/getting-started-with-embeddings\n",
    "# https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2\n",
    "embeddings=HuggingFaceEmbeddings(model_name='sentence-transformers/all-MiniLM-L6-v2')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Generate the embeddings for the documents and store them in the Chroma Vector Store.\n",
    "db = Chroma.from_documents(documents, embeddings, persist_directory=\"./chroma_db/\")\n",
    "db.persist()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "llama2",
   "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": "3.11.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}