Spaces:
Running
Running
File size: 6,095 Bytes
364893a 62e78cf 364893a |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 |
{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Used this to migrate vectors to pinecone from our faiss indices. I recommend you use our scripts to ingest your data directly into Pinecone. For this, direct it to a folder containing the index.faiss and index.pkl files that you want to ingest into pinecone."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"c:\\Users\\dfole\\Desktop\\CS549\\pinecone_venv\\Lib\\site-packages\\pinecone\\data\\index.py:1: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html\n",
" from tqdm.autonotebook import tqdm\n"
]
}
],
"source": [
"import getpass\n",
"import os\n",
"import time\n",
"from pinecone import Pinecone, ServerlessSpec\n",
"\n",
"pinecone_api_key = os.environ.get(\"PINECONE_API_KEY\")\n",
"\n",
"pc = Pinecone(api_key=pinecone_api_key)"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {},
"outputs": [],
"source": [
"from langchain_huggingface import HuggingFaceEmbeddings\n",
"\n",
"embeddings = HuggingFaceEmbeddings(model_name=\"sentence-transformers/all-MiniLM-L6-v2\")"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|██████████| 4685/4685 [1:57:28<00:00, 1.50s/it] \n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Successfully migrated 468455 documents to Pinecone index 'bpl-rag'\n"
]
}
],
"source": [
"import os\n",
"from langchain_community.vectorstores import FAISS\n",
"from pinecone import Pinecone, ServerlessSpec\n",
"from langchain_community.embeddings import OpenAIEmbeddings\n",
"from tqdm import tqdm\n",
"from langchain_pinecone import PineconeVectorStore\n",
"\n",
"def migrate_faiss_to_pinecone(\n",
" faiss_index_path: str,\n",
" pinecone_api_key: str,\n",
" index_name: str,\n",
" batch_size: int = 100\n",
"):\n",
" \"\"\"\n",
" Migrate a local FAISS index to Pinecone.\n",
" \n",
" Args:\n",
" faiss_index_path: Path to the local FAISS index\n",
" pinecone_api_key: Your Pinecone API key\n",
" pinecone_environment: Pinecone environment (e.g., \"us-east1-gcp\")\n",
" index_name: Name of the Pinecone index to create/use\n",
" batch_size: Number of vectors to upload in each batch\n",
" \"\"\"\n",
" # Load the local FAISS index\n",
" embeddings = HuggingFaceEmbeddings(model_name=\"sentence-transformers/all-MiniLM-L6-v2\")\n",
" faiss_vectorstore = FAISS.load_local(faiss_index_path, embeddings,allow_dangerous_deserialization=True)\n",
" pc = Pinecone(api_key=pinecone_api_key)\n",
"\n",
" index = pc.Index(index_name)\n",
" \n",
" # Get all the vectors and documents from FAISS\n",
" all_docs = faiss_vectorstore.docstore._dict\n",
" docs = dict()\n",
"\n",
" for uuid in faiss_vectorstore.docstore._dict:\n",
" doc = faiss_vectorstore.docstore._dict[uuid]\n",
" # print(doc)\n",
" if doc.metadata['field'] in ['abstract_tsi','title_info_primary_tsi','title_info_primary_subtitle_tsi', 'title_info_alternative_tsim']:\n",
" if len(doc.page_content) > 3:\n",
" docs[uuid] = doc\n",
"\n",
" total_docs = len(docs)\n",
" \n",
" pinecone_vectorstore = PineconeVectorStore(index=index, embedding=embeddings)\n",
"\n",
" # Batch processing\n",
" for i in tqdm(range(0, total_docs, batch_size)):\n",
" batch_ids = list(docs.keys())[i:i + batch_size]\n",
" batch_docs = [docs[doc_id] for doc_id in batch_ids]\n",
" batch_embeddings = [faiss_vectorstore.index.reconstruct(j).tolist() \n",
" for j in range(i, min(i + batch_size, total_docs))]\n",
" \n",
" # Create metadata for each document\n",
" metadatas = [doc.metadata for doc in batch_docs]\n",
" texts = [doc.page_content for doc in batch_docs]\n",
" # print(batch_docs)\n",
" # Add vectors to Pinecone\n",
" pinecone_vectorstore.add_texts(\n",
" texts=texts,\n",
" metadatas=metadatas,\n",
" embeddings=batch_embeddings,\n",
" ids=batch_ids\n",
" )\n",
" \n",
" print(f\"Successfully migrated {total_docs} documents to Pinecone index '{index_name}'\")\n",
" return pinecone_vectorstore\n",
"\n",
"# Example usage:\n",
"if __name__ == \"__main__\":\n",
" # Set your credentials and paths\n",
" FAISS_INDEX_PATH = \"faiss_900_1200\"\n",
" PINECONE_API_KEY = \os.get.environ("PINECONE_API_KEY"),
" INDEX_NAME = \"bpl-rag\"\n",
" \n",
" # Perform migration\n",
" pinecone_vs = migrate_faiss_to_pinecone(\n",
" faiss_index_path=FAISS_INDEX_PATH,\n",
" pinecone_api_key=PINECONE_API_KEY,\n",
" index_name=INDEX_NAME,\n",
" batch_size=100\n",
" )"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
}
],
"metadata": {
"kernelspec": {
"display_name": "pinecone_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": "3.12.4"
}
},
"nbformat": 4,
"nbformat_minor": 2
}
|