Spaces:
Sleeping
Sleeping
File size: 9,549 Bytes
4e5fc16 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 |
"""Vector database management for Francis Botcon."""
import json
from pathlib import Path
from typing import List, Dict, Tuple
import numpy as np
from src.embeddings import EmbeddingGenerator
from src.logger import LoggerSetup
from src.config_loader import config
logger = LoggerSetup.setup().getChild(__name__)
class VectorDatabase:
"""Manage vector embeddings and retrieval using ChromaDB or FAISS."""
def __init__(self, db_type: str = None, db_path: str = None):
"""Initialize vector database.
Args:
db_type: Type of database ('chromadb' or 'faiss')
db_path: Path to database
"""
self.db_type = db_type or config.get("vector_db.type", "chromadb")
self.db_path = Path(db_path or config.get("vector_db.db_path", "./data/vectordb"))
self.db_path.mkdir(parents=True, exist_ok=True)
self.embedding_generator = EmbeddingGenerator()
self.top_k = config.get("vector_db.top_k", 5)
self.similarity_threshold = config.get("vector_db.similarity_threshold", 0.6)
logger.info(f"Initializing {self.db_type} database at {self.db_path}")
if self.db_type == "chromadb":
self._init_chromadb()
elif self.db_type == "faiss":
self._init_faiss()
else:
raise ValueError(f"Unsupported database type: {self.db_type}")
def _init_chromadb(self):
"""Initialize ChromaDB."""
try:
import chromadb
self.client = chromadb.PersistentClient(path=str(self.db_path))
self.collection = None
# Try to load existing collection
try:
self.collection = self.client.get_collection(name="francis_bacon")
logger.info("✓ ChromaDB initialized - loaded existing collection")
except Exception as e:
logger.debug(f"No existing collection found: {e}. Will create on first add_documents call.")
logger.info("✓ ChromaDB initialized")
except ImportError:
logger.error("ChromaDB not installed. Install with: pip install chromadb")
raise
def _init_faiss(self):
"""Initialize FAISS."""
try:
import faiss
self.faiss = faiss
self.index = None
self.documents = []
logger.info("✓ FAISS initialized")
except ImportError:
logger.error("FAISS not installed. Install with: pip install faiss-cpu")
raise
def add_documents(self, documents: List[Dict[str, str]], batch_size: int = 32):
"""Add documents to vector database.
Args:
documents: List of documents with 'id', 'text', and metadata
batch_size: Batch size for embedding generation
"""
logger.info(f"Adding {len(documents)} documents to {self.db_type} database")
# Extract texts for embedding
texts = [doc["text"] for doc in documents]
# Generate embeddings
embeddings = self.embedding_generator.embed(texts, batch_size=batch_size)
if self.db_type == "chromadb":
self._add_to_chromadb(documents, embeddings, texts)
elif self.db_type == "faiss":
self._add_to_faiss(documents, embeddings, texts)
logger.info("✓ Documents added successfully")
def _add_to_chromadb(self, documents: List[Dict], embeddings: np.ndarray, texts: List[str]):
"""Add documents to ChromaDB.
Args:
documents: Document list
embeddings: Embedding vectors
texts: Text strings
"""
# Create collection if not exists
if self.collection is None:
self.collection = self.client.get_or_create_collection(
name="francis_bacon",
metadata={"hnsw:space": "cosine"}
)
# Prepare metadata
metadatas = []
ids = []
for i, doc in enumerate(documents):
ids.append(doc["id"])
metadatas.append({
"source": doc.get("source", ""),
"title": doc.get("title", ""),
"author": doc.get("author", ""),
"segment_index": str(doc.get("segment_index", 0))
})
# Add to collection
self.collection.add(
ids=ids,
embeddings=embeddings.tolist(),
documents=texts,
metadatas=metadatas
)
def _add_to_faiss(self, documents: List[Dict], embeddings: np.ndarray, texts: List[str]):
"""Add documents to FAISS.
Args:
documents: Document list
embeddings: Embedding vectors
texts: Text strings
"""
# Initialize index if needed
if self.index is None:
embedding_dim = embeddings.shape[1]
self.index = self.faiss.IndexFlatL2(embedding_dim)
# Convert to float32 for FAISS
embeddings_float32 = embeddings.astype(np.float32)
# Add vectors
self.index.add(embeddings_float32)
# Store documents
for doc, text in zip(documents, texts):
doc["embedding_index"] = len(self.documents)
self.documents.append({**doc, "text": text})
# Save index
self._save_faiss_index()
def search(self, query: str, top_k: int = None) -> List[Tuple[str, float, Dict]]:
"""Search for similar documents.
Args:
query: Query text
top_k: Number of results to return
Returns:
List of (text, score, metadata) tuples
"""
top_k = top_k or self.top_k
# Generate query embedding
query_embedding = self.embedding_generator.embed_single(query)
if self.db_type == "chromadb":
return self._search_chromadb(query_embedding, top_k)
elif self.db_type == "faiss":
return self._search_faiss(query_embedding, top_k)
def _search_chromadb(self, query_embedding: np.ndarray, top_k: int) -> List[Tuple[str, float, Dict]]:
"""Search ChromaDB.
Args:
query_embedding: Query embedding vector
top_k: Number of results
Returns:
Search results
"""
results = self.collection.query(
query_embeddings=[query_embedding.tolist()],
n_results=top_k,
include=["documents", "distances", "metadatas"]
)
output = []
if results["documents"] and len(results["documents"]) > 0:
for i, doc in enumerate(results["documents"][0]):
# ChromaDB uses distance, convert to similarity (cosine)
distance = results["distances"][0][i]
similarity = 1 - (distance / 2) # Approximate cosine conversion
metadata = results["metadatas"][0][i] if results["metadatas"] else {}
if similarity >= self.similarity_threshold:
output.append((doc, similarity, metadata))
return output
def _search_faiss(self, query_embedding: np.ndarray, top_k: int) -> List[Tuple[str, float, Dict]]:
"""Search FAISS.
Args:
query_embedding: Query embedding vector
top_k: Number of results
Returns:
Search results
"""
query_embedding_float32 = query_embedding.astype(np.float32).reshape(1, -1)
distances, indices = self.index.search(query_embedding_float32, top_k)
output = []
for i, idx in enumerate(indices[0]):
if idx != -1:
# Convert L2 distance to similarity
distance = distances[0][i]
similarity = 1 / (1 + distance)
if similarity >= self.similarity_threshold:
doc_info = self.documents[idx]
metadata = {
"source": doc_info.get("source", ""),
"title": doc_info.get("title", ""),
"author": doc_info.get("author", "")
}
output.append((doc_info["text"], similarity, metadata))
return output
def _save_faiss_index(self):
"""Save FAISS index and documents."""
if self.db_type == "faiss":
import faiss
index_path = self.db_path / "faiss_index.bin"
docs_path = self.db_path / "documents.json"
faiss.write_index(self.index, str(index_path))
with open(docs_path, 'w') as f:
json.dump(self.documents, f, ensure_ascii=False, indent=2)
logger.debug(f"FAISS index saved to {index_path}")
def load_index(self):
"""Load existing FAISS index."""
if self.db_type == "faiss":
import faiss
index_path = self.db_path / "faiss_index.bin"
docs_path = self.db_path / "documents.json"
if index_path.exists() and docs_path.exists():
self.index = faiss.read_index(str(index_path))
with open(docs_path, 'r') as f:
self.documents = json.load(f)
logger.info("✓ FAISS index loaded")
return True
return False
|