Spaces:
Sleeping
Sleeping
File size: 3,748 Bytes
cdb5c98 e7ec62f cdb5c98 e7ec62f cdb5c98 e7ec62f c2bd4e0 e7ec62f c2bd4e0 e7ec62f cdb5c98 e7ec62f c2bd4e0 e7ec62f cdb5c98 c2bd4e0 cdb5c98 e7ec62f cdb5c98 c2bd4e0 e7ec62f cdb5c98 e7ec62f c2bd4e0 e7ec62f cdb5c98 e7ec62f cdb5c98 c2bd4e0 cdb5c98 e7ec62f cdb5c98 e7ec62f cdb5c98 c2bd4e0 cdb5c98 e7ec62f c2bd4e0 e7ec62f c2bd4e0 e7ec62f c2bd4e0 e7ec62f c2bd4e0 e7ec62f c2bd4e0 e7ec62f c2bd4e0 e7ec62f | 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 | import faiss
import numpy as np
import pickle
import os
class VectorStore:
def __init__(
self,
index_path="data/vector.index",
docs_path="data/documents.pkl"
):
self.index = None
self.documents = []
self.index_path = index_path
self.docs_path = docs_path
# Load existing database
self.load()
# =====================================================
# BUILD VECTOR DATABASE
# =====================================================
def build(
self,
embeddings,
documents
):
"""
Create FAISS vector database.
embeddings:
SentenceTransformer embeddings
documents:
text chunks
"""
if len(embeddings) == 0:
return
embeddings = np.array(
embeddings
).astype("float32")
# Normalize for cosine similarity
faiss.normalize_L2(
embeddings
)
dimension = embeddings.shape[1]
# Cosine similarity search
self.index = faiss.IndexFlatIP(
dimension
)
self.index.add(
embeddings
)
self.documents = documents
self.save()
# =====================================================
# SEARCH
# =====================================================
def search(
self,
query_embedding,
k=5
):
if self.index is None:
return []
query_embedding = np.array(
[query_embedding]
).astype("float32")
faiss.normalize_L2(
query_embedding
)
distances, indices = self.index.search(
query_embedding,
k
)
results = []
for score, idx in zip(
distances[0],
indices[0]
):
if idx != -1:
results.append(
self.documents[idx]
)
return results
# =====================================================
# SAVE DATABASE
# =====================================================
def save(self):
"""
Save FAISS index + documents.
Creates folders automatically.
"""
# Create directories if missing
index_dir = os.path.dirname(
self.index_path
)
docs_dir = os.path.dirname(
self.docs_path
)
if index_dir:
os.makedirs(
index_dir,
exist_ok=True
)
if docs_dir:
os.makedirs(
docs_dir,
exist_ok=True
)
# Save FAISS index
if self.index is not None:
faiss.write_index(
self.index,
self.index_path
)
# Save documents
with open(
self.docs_path,
"wb"
) as f:
pickle.dump(
self.documents,
f
)
# =====================================================
# LOAD DATABASE
# =====================================================
def load(self):
"""
Load FAISS database if available.
"""
if os.path.exists(
self.index_path
):
self.index = faiss.read_index(
self.index_path
)
if os.path.exists(
self.docs_path
):
with open(
self.docs_path,
"rb"
) as f:
self.documents = pickle.load(f) |