forked from lancedb/lancedb
-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathimagebind.py
More file actions
182 lines (159 loc) · 6.25 KB
/
imagebind.py
File metadata and controls
182 lines (159 loc) · 6.25 KB
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
# Copyright (c) 2023. LanceDB Developers
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
from functools import cached_property
from typing import List, Union
import numpy as np
import pyarrow as pa
from ..util import attempt_import_or_raise
from .base import EmbeddingFunction
from .registry import register
from .utils import AUDIO, IMAGES, TEXT
from lancedb.pydantic import PYDANTIC_VERSION
@register("imagebind")
class ImageBindEmbeddings(EmbeddingFunction):
"""
An embedding function that uses the ImageBind API
For generating multi-modal embeddings across
six different modalities: images, text, audio, depth, thermal, and IMU data
to download package, run :
`pip install imagebind-packaged==0.1.2`
"""
name: str = "imagebind_huge"
device: str = "cpu"
normalize: bool = False
if PYDANTIC_VERSION.major < 2: # Pydantic 1.x compat
class Config:
keep_untouched = (cached_property,)
else:
model_config = dict()
model_config["ignored_types"] = (cached_property,)
def __init__(self, *args, **kwargs):
super().__init__(*args, **kwargs)
self._ndims = 1024
self._audio_extensions = (".mp3", ".wav", ".flac", ".ogg", ".aac")
self._image_extensions = (".jpg", ".jpeg", ".png", ".gif", ".bmp")
@cached_property
def embedding_model(self):
"""
Get the embedding model. This is cached so that the model is only loaded
once per process.
"""
return self.get_embedding_model()
@cached_property
def _data(self):
"""
Get the data module from imagebind
"""
data = attempt_import_or_raise("imagebind.data", "imagebind")
return data
@cached_property
def _ModalityType(self):
"""
Get the ModalityType from imagebind
"""
imagebind = attempt_import_or_raise("imagebind", "imagebind")
return imagebind.imagebind_model.ModalityType
def ndims(self):
return self._ndims
def compute_query_embeddings(
self, query: Union[str], *args, **kwargs
) -> List[np.ndarray]:
"""
Compute the embeddings for a given user query
Parameters
----------
query : Union[str]
The query to embed. A query can be either text, image paths or audio paths.
"""
query = self.sanitize_input(query)
if query[0].endswith(self._audio_extensions):
return [self.generate_audio_embeddings(query)]
elif query[0].endswith(self._image_extensions):
return [self.generate_image_embeddings(query)]
else:
return [self.generate_text_embeddings(query)]
def generate_image_embeddings(self, image: IMAGES) -> np.ndarray:
torch = attempt_import_or_raise("torch")
inputs = {
self._ModalityType.VISION: self._data.load_and_transform_vision_data(
image, self.device
)
}
with torch.no_grad():
image_features = self.embedding_model(inputs)[self._ModalityType.VISION]
if self.normalize:
image_features /= image_features.norm(dim=-1, keepdim=True)
return image_features.cpu().numpy().squeeze()
def generate_audio_embeddings(self, audio: AUDIO) -> np.ndarray:
torch = attempt_import_or_raise("torch")
inputs = {
self._ModalityType.AUDIO: self._data.load_and_transform_audio_data(
audio, self.device
)
}
with torch.no_grad():
audio_features = self.embedding_model(inputs)[self._ModalityType.AUDIO]
if self.normalize:
audio_features /= audio_features.norm(dim=-1, keepdim=True)
return audio_features.cpu().numpy().squeeze()
def generate_text_embeddings(self, text: TEXT) -> np.ndarray:
torch = attempt_import_or_raise("torch")
inputs = {
self._ModalityType.TEXT: self._data.load_and_transform_text(
text, self.device
)
}
with torch.no_grad():
text_features = self.embedding_model(inputs)[self._ModalityType.TEXT]
if self.normalize:
text_features /= text_features.norm(dim=-1, keepdim=True)
return text_features.cpu().numpy().squeeze()
def compute_source_embeddings(
self, source: Union[IMAGES, AUDIO], *args, **kwargs
) -> List[np.array]:
"""
Get the embeddings for the given sourcefield column in the pydantic model.
"""
source = self.sanitize_input(source)
embeddings = []
if source[0].endswith(self._audio_extensions):
embeddings.extend(self.generate_audio_embeddings(source))
return embeddings
elif source[0].endswith(self._image_extensions):
embeddings.extend(self.generate_image_embeddings(source))
return embeddings
else:
embeddings.extend(self.generate_text_embeddings(source))
return embeddings
def sanitize_input(
self, input: Union[IMAGES, AUDIO]
) -> Union[List[bytes], np.ndarray]:
"""
Sanitize the input to the embedding function.
"""
if isinstance(input, (str, bytes)):
input = [input]
elif isinstance(input, pa.Array):
input = input.to_pylist()
elif isinstance(input, pa.ChunkedArray):
input = input.combine_chunks().to_pylist()
return input
def get_embedding_model(self):
"""
fetches the imagebind embedding model
"""
imagebind = attempt_import_or_raise("imagebind", "imagebind")
model = imagebind.imagebind_model.imagebind_huge(pretrained=True)
model.eval()
model.to(self.device)
return model