Spaces:
Running
Running
File size: 10,996 Bytes
b5d3a91 | 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 | # built-in dependencies
import os
import json
import hashlib
import struct
import base64
import uuid
import math
from typing import Any, Dict, Optional, List, Union
# project dependencies
from deepface.modules.database.types import Database
from deepface.commons.logger import Logger
logger = Logger()
_SCHEMA_CHECKED: Dict[str, bool] = {}
# pylint: disable=too-many-positional-arguments
class WeaviateClient(Database):
"""
Weaviate client for storing and retrieving face embeddings and indices.
"""
def __init__(
self,
connection_details: Optional[Union[str, Dict[str, Any]]] = None,
connection: Any = None,
):
try:
import weaviate
except (ModuleNotFoundError, ImportError) as e:
raise ValueError(
"weaviate-client is an optional dependency. "
"Install with 'pip install weaviate-client'"
) from e
self.weaviate = weaviate
if connection is not None:
self.client = connection
# URL key for _WEAVIATE_CHECKED; fallback if client has no URL
self.url = getattr(connection, "url", str(id(connection)))
else:
self.conn_details = connection_details or os.environ.get("DEEPFACE_WEAVIATE_URL")
if isinstance(self.conn_details, str):
self.url = self.conn_details
self.api_key = os.getenv("WEAVIATE_API_KEY")
elif isinstance(self.conn_details, dict):
self.url = self.conn_details.get("url")
self.api_key = self.conn_details.get("api_key") or os.getenv("WEAVIATE_API_KEY")
else:
raise ValueError("connection_details must be a string or dict with 'url'.")
if not self.url:
raise ValueError("Weaviate URL not provided in connection_details.")
client_config = {"url": self.url}
if getattr(self, "api_key", None):
client_config["auth_client_secret"] = self.weaviate.AuthApiKey(api_key=self.api_key)
self.client = self.weaviate.Client(**client_config)
def initialize_database(self, **kwargs: Any) -> None:
"""
Ensure Weaviate schemas exist for embeddings using both cosine and L2 (euclidean).
"""
model_name = kwargs.get("model_name", "VGG-Face")
detector_backend = kwargs.get("detector_backend", "opencv")
aligned = kwargs.get("aligned", True)
l2_normalized = kwargs.get("l2_normalized", False)
existing_schema = self.client.schema.get()
existing_classes = {c["class"] for c in existing_schema.get("classes", [])}
class_name = self.__generate_class_name(
model_name=model_name,
detector_backend=detector_backend,
aligned=aligned,
l2_normalized=l2_normalized,
)
if _SCHEMA_CHECKED.get(class_name):
logger.debug("Weaviate schema already checked, skipping.")
return
if class_name in existing_classes:
logger.debug(f"Weaviate class {class_name} already exists.")
return
self.client.schema.create_class(
{
"class": class_name,
"vectorIndexType": "hnsw",
"vectorizer": "none",
"vectorIndexConfig": {
"M": int(os.getenv("WEAVIATE_HNSW_M", "16")),
"distance": "cosine" if l2_normalized else "l2-squared",
},
"properties": [
{"name": "img_name", "dataType": ["text"]},
{"name": "face", "dataType": ["blob"]},
{"name": "face_shape", "dataType": ["int[]"]},
{"name": "model_name", "dataType": ["text"]},
{"name": "detector_backend", "dataType": ["text"]},
{"name": "aligned", "dataType": ["boolean"]},
{"name": "l2_normalized", "dataType": ["boolean"]},
{"name": "face_hash", "dataType": ["text"]},
{"name": "embedding_hash", "dataType": ["text"]},
# embedding property is optional since we pass it as vector
{"name": "embedding", "dataType": ["number[]"]},
],
}
)
logger.debug(f"Weaviate class {class_name} created successfully.")
_SCHEMA_CHECKED[class_name] = True
def insert_embeddings(self, embeddings: List[Dict[str, Any]], batch_size: int = 100) -> int:
"""
Insert multiple embeddings into Weaviate using batch API.
"""
if not embeddings:
raise ValueError("No embeddings to insert.")
self.initialize_database(
model_name=embeddings[0]["model_name"],
detector_backend=embeddings[0]["detector_backend"],
aligned=embeddings[0]["aligned"],
l2_normalized=embeddings[0]["l2_normalized"],
)
class_name = self.__generate_class_name(
model_name=embeddings[0]["model_name"],
detector_backend=embeddings[0]["detector_backend"],
aligned=embeddings[0]["aligned"],
l2_normalized=embeddings[0]["l2_normalized"],
)
with self.client.batch as batcher:
batcher.batch_size = batch_size
batcher.timeout_retries = 3
for e in embeddings:
face_json = json.dumps(e["face"].tolist())
face_hash = hashlib.sha256(face_json.encode()).hexdigest()
embedding_bytes = struct.pack(f'{len(e["embedding"])}d', *e["embedding"])
embedding_hash = hashlib.sha256(embedding_bytes).hexdigest()
# Check if embedding already exists
query = (
self.client.query.get(class_name, ["embedding_hash"])
.with_where(
{
"path": ["embedding_hash"],
"operator": "Equal",
"valueText": embedding_hash,
}
)
.with_limit(1)
.do()
)
existing = query.get("data", {}).get("Get", {}).get(class_name, [])
if existing:
logger.warn(
f"Embedding with hash {embedding_hash} already exists in {class_name}."
)
continue
uid = str(uuid.uuid4())
properties = {
"img_name": e["img_name"],
"face": base64.b64encode(e["face"].tobytes()).decode("utf-8"),
"face_shape": list(e["face"].shape),
"model_name": e["model_name"],
"detector_backend": e["detector_backend"],
"aligned": e["aligned"],
"l2_normalized": e["l2_normalized"],
"embedding": e["embedding"], # optional
"face_hash": face_hash,
"embedding_hash": embedding_hash,
}
batcher.add_data_object(properties, class_name, vector=e["embedding"], uuid=uid)
return len(embeddings)
def fetch_all_embeddings(
self,
model_name: str,
detector_backend: str,
aligned: bool,
l2_normalized: bool,
batch_size: int = 1000,
) -> List[Dict[str, Any]]:
"""
Fetch all embeddings with filters.
"""
class_name = self.__generate_class_name(
model_name=model_name,
detector_backend=detector_backend,
aligned=aligned,
l2_normalized=l2_normalized,
)
self.initialize_database(
model_name=model_name,
detector_backend=detector_backend,
aligned=aligned,
l2_normalized=l2_normalized,
)
results = (
self.client.query.get(class_name, ["img_name", "embedding"])
.with_additional(["id"])
.do()
)
data = results.get("data", {}).get("Get", {}).get(class_name, [])
embeddings = []
for r in data:
embeddings.append(
{
"id": r.get("_additional", {}).get("id"),
"img_name": r["img_name"],
"embedding": r["embedding"],
"model_name": model_name,
"detector_backend": detector_backend,
"aligned": aligned,
"l2_normalized": l2_normalized,
}
)
return embeddings
def search_by_vector(
self,
vector: List[float],
model_name: str = "VGG-Face",
detector_backend: str = "opencv",
aligned: bool = True,
l2_normalized: bool = False,
limit: int = 10,
) -> List[Dict[str, Any]]:
"""
ANN search using the main vector (embedding).
"""
class_name = self.__generate_class_name(
model_name=model_name,
detector_backend=detector_backend,
aligned=aligned,
l2_normalized=l2_normalized,
)
self.initialize_database(
model_name=model_name,
detector_backend=detector_backend,
aligned=aligned,
l2_normalized=l2_normalized,
)
query = self.client.query.get(class_name, ["img_name", "embedding"])
query = (
query.with_near_vector({"vector": vector})
.with_limit(limit)
.with_additional(["id", "distance"])
)
results = query.do()
data = results.get("data", {}).get("Get", {}).get(class_name, [])
return [
{
"id": r.get("_additional", {}).get("id"),
"img_name": r["img_name"],
"embedding": r["embedding"],
"distance": (
r.get("_additional", {}).get("distance")
if l2_normalized
else math.sqrt(r.get("_additional", {}).get("distance"))
),
}
for r in data
]
def close(self) -> None:
"""
Close the Weaviate client connection.
"""
self.client.close()
@staticmethod
def __generate_class_name(
model_name: str,
detector_backend: str,
aligned: bool,
l2_normalized: bool,
) -> str:
"""
Generate Weaviate class name based on parameters.
"""
class_name_attributes = [
model_name.replace("-", ""),
detector_backend,
"Aligned" if aligned else "Unaligned",
"Norm" if l2_normalized else "Raw",
]
return "Embeddings_" + "_".join(class_name_attributes).lower()
|