{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- Write a Python notebook that does semantic search on the vector database and return top k results (use LangChain). Comment on what you observe."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sentence_transformers import SentenceTransformer\n",
    "import numpy as np\n",
    "from tqdm import tqdm\n",
    "import os\n",
    "from langchain.vectorstores import Chroma"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Wrapper with embed_documents and embed_query\n",
    "class SentenceTransformerWrapper:\n",
    "    def __init__(self, model_name):\n",
    "        self.model = SentenceTransformer(model_name)\n",
    "        \n",
    "    def embed_documents(self, texts):\n",
    "        # Convert the list of texts to embeddings\n",
    "        return self.model.encode(texts, show_progress_bar=True).tolist()\n",
    "    \n",
    "    def embed_query(self, text):\n",
    "        # Convert a single query to its embedding\n",
    "        return self.model.encode(text).tolist()\n",
    "\n",
    "# Instantiate wrapper with model\n",
    "embedding_model = SentenceTransformerWrapper('bkai-foundation-models/vietnamese-bi-encoder')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Chroma database\n",
    "vector_db = Chroma(\n",
    "    persist_directory=\"chroma_db_new\",\n",
    "    embedding=embedding_model  # Use your SentenceTransformerWrapper instance\n",
    ")\n",
    "\n",
    "# Test by running a similarity search\n",
    "query = input(\"Enter your query: \")\n",
    "results = vector_db.similarity_search(query, k=5)\n",
    "\n",
    "# Display the results\n",
    "print(f\"\\nTop 5 results for query: '{query}'\\n\")\n",
    "for i, doc in enumerate(results):\n",
    "    print(f\"Result {i+1}:\")\n",
    "    print(f\"Metadata: {doc.metadata}\")\n",
    "    print(f\"Content: {doc.page_content[:50]}...\")  # Display a preview of the chunk\n",
    "    print(\"-\" * 50)\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "phapdienvv",
   "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": "3.11.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}