Spaces:
Running
Running
File size: 11,180 Bytes
0a4529c |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 |
# DEPENDENCIES
import numpy as np
from typing import List
from typing import Optional
from config.models import ChunkWithScore
from config.models import DocumentChunk
from config.settings import get_settings
from config.logging_config import get_logger
from utils.error_handler import handle_errors
from embeddings.bge_embedder import get_embedder
from utils.error_handler import VectorSearchError
from vector_store.faiss_manager import get_faiss_manager
from vector_store.metadata_store import get_metadata_store
# Setup Settings and Logging
settings = get_settings()
logger = get_logger(__name__)
class VectorSearch:
"""
FAISS-based vector similarity search: Uses existing FAISSManager from vector_store module
Performs semantic search using embedding similarity
"""
def __init__(self):
"""
Initialize vector search
"""
self.logger = logger
self.faiss_manager = get_faiss_manager()
self.embedder = get_embedder()
self.metadata_store = get_metadata_store()
# Search statistics
self.search_count = 0
self.total_results = 0
self.logger.info("Initialized VectorSearch")
@handle_errors(error_type = VectorSearchError, log_error = True, reraise = True)
def search(self, query: str, top_k: int = 10, min_score: float = 0.0) -> List[ChunkWithScore]:
"""
Perform vector similarity search
Arguments:
----------
query { str } : Search query
top_k { int } : Number of results to return
min_score { float } : Minimum similarity score threshold
Returns:
--------
{ list } : List of ChunkWithScore objects
"""
if not query or not query.strip():
self.logger.warning("Empty query provided to vector search")
return []
self.logger.debug(f"Performing vector search: '{query}' (top_k={top_k})")
try:
# Generate query embedding
query_embedding = self.embedder.embed_text(text = query,
normalize = True,
)
# Search FAISS index (returns List[Tuple[str, float]] = [(chunk_id, score), ...])
faiss_results = self.faiss_manager.search(query_embedding = query_embedding,
top_k = top_k,
)
if not faiss_results:
self.logger.info(f"No results found for query: '{query}'")
return []
# Convert to ChunkWithScore objects
chunks_with_scores = list()
for rank, (chunk_id, score) in enumerate(faiss_results, 1):
# Filter by minimum score
if (score < min_score):
continue
# Get chunk metadata
chunk_metadata = self.metadata_store.get_chunk_metadata(chunk_id)
if not chunk_metadata:
self.logger.warning(f"Chunk metadata not found for: {chunk_id}")
continue
# Create DocumentChunk
chunk = self._metadata_to_chunk(chunk_metadata)
# Create ChunkWithScore
cws = ChunkWithScore(chunk = chunk,
score = score,
rank = rank,
retrieval_method = 'vector',
)
chunks_with_scores.append(cws)
# Update statistics
self.search_count += 1
self.total_results += len(chunks_with_scores)
self.logger.info(f"Vector search returned {len(chunks_with_scores)} results")
return chunks_with_scores
except Exception as e:
self.logger.error(f"Vector search failed: {repr(e)}")
raise VectorSearchError(f"Vector search failed: {repr(e)}")
def search_with_embedding(self, query_embedding: np.ndarray, top_k: int = 10, min_score: float = 0.0) -> List[ChunkWithScore]:
"""
Search using pre-computed query embedding
Arguments:
----------
query_embedding { np.ndarray } : Query embedding vector
top_k { int } : Number of results
min_score { float } : Minimum score threshold
Returns:
--------
{ list } : List of ChunkWithScore objects
"""
self.logger.debug(f"Performing vector search with pre-computed embedding (top_k={top_k})")
try:
# Search FAISS index
faiss_results = self.faiss_manager.search(query_embedding=query_embedding, top_k=top_k)
# Convert to ChunkWithScore objects
chunks_with_scores = list()
for rank, (chunk_id, score) in enumerate(faiss_results, 1):
if (score < min_score):
continue
chunk_metadata = self.metadata_store.get_chunk_metadata(chunk_id)
if not chunk_metadata:
continue
chunk = self._metadata_to_chunk(chunk_metadata)
cws = ChunkWithScore(chunk = chunk,
score = score,
rank = rank,
retrieval_method = 'vector',
)
chunks_with_scores.append(cws)
self.search_count += 1
self.total_results += len(chunks_with_scores)
return chunks_with_scores
except Exception as e:
self.logger.error(f"Vector search with embedding failed: {repr(e)}")
raise VectorSearchError(f"Vector search with embedding failed: {repr(e)}")
def _metadata_to_chunk(self, metadata: dict) -> DocumentChunk:
"""
Convert metadata dictionary to DocumentChunk object
Arguments:
----------
metadata { dict } : Chunk metadata from store
Returns:
--------
{ DocumentChunk } : DocumentChunk object
"""
return DocumentChunk(chunk_id = metadata['chunk_id'],
document_id = metadata['document_id'],
text = metadata['text'],
embedding = metadata.get('embedding'),
chunk_index = metadata['chunk_index'],
start_char = metadata['start_char'],
end_char = metadata['end_char'],
page_number = metadata.get('page_number'),
section_title = metadata.get('section_title'),
token_count = metadata['token_count'],
metadata = metadata.get('metadata', {}),
)
def search_with_filters(self, query: str, top_k: int = 10, document_ids: Optional[List[str]] = None,
min_score: float = 0.0) -> List[ChunkWithScore]:
"""
Search with document filters
Arguments:
----------
query { str } : Search query
top_k { int } : Number of results
document_ids { list } : Filter by specific documents
min_score { float } : Minimum score threshold
Returns:
--------
{ list } : Filtered ChunkWithScore objects
"""
# Get more results for filtering
results = self.search(query = query,
top_k = top_k * 2,
min_score = min_score,
)
# Filter by document IDs if provided
if document_ids:
results = [r for r in results if r.chunk.document_id in document_ids]
# Return top_k after filtering
return results[:top_k]
def batch_search(self, queries: List[str], top_k: int = 10) -> List[List[ChunkWithScore]]:
"""
Perform batch vector search for multiple queries
Arguments:
----------
queries { list } : List of query strings
top_k { int } : Number of results per query
Returns:
--------
{ list } : List of result lists
"""
self.logger.info(f"Performing batch vector search for {len(queries)} queries")
results = list()
for query in queries:
query_results = self.search(query, top_k)
results.append(query_results)
return results
def get_search_statistics(self) -> dict:
"""
Get vector search statistics
Returns:
--------
{ dict } : Search statistics
"""
avg_results = (self.total_results / self.search_count) if (self.search_count > 0) else 0
return {"search_count" : self.search_count,
"total_results" : self.total_results,
"avg_results_per_query" : avg_results,
"faiss_index_stats" : self.faiss_manager.get_index_stats(),
"embedding_model" : self.embedder.model_name,
"embedding_dimension" : self.embedder.embedding_dim,
}
# Global vector search instance
_vector_search = None
def get_vector_search() -> VectorSearch:
"""
Get global vector search instance
Returns:
--------
{ VectorSearch } : VectorSearch instance
"""
global _vector_search
if _vector_search is None:
_vector_search = VectorSearch()
return _vector_search
def search_vectors(query: str, top_k: int = 10, **kwargs) -> List[ChunkWithScore]:
"""
Convenience function for vector search
Arguments:
----------
query { str } : Search query
top_k { int } : Number of results
**kwargs : Additional arguments
Returns:
--------
{ list } : ChunkWithScore results
"""
searcher = get_vector_search()
return searcher.search(query, top_k, **kwargs) |