Hugging Face's logo

Xenova
/
slimsam-77-uniform

Mask Generation
Transformers.js
ONNX
sam
Model card Files Community
2
Xenova HF staff commited on
Commit
4050e09
•
1 Parent(s): 39b5ccf

Upload folder using huggingface_hub

Browse files
Files changed (7) hide show
  1. config.json +1 -1
  2. onnx/prompt_encoder_mask_decoder.onnx +2 -2
  3. onnx/prompt_encoder_mask_decoder_quantized.onnx +2 -2
  4. onnx/vision_encoder.onnx +2 -2
  5. onnx/vision_encoder_quantized.onnx +2 -2
  6. preprocessor_config.json +7 -0
  7. quantize_config.json +55 -57
config.json CHANGED
@@ -11,7 +11,7 @@
11
  "prompt_encoder_config": {
12
  "model_type": ""
13
  },
14
- "transformers_version": "4.36.2",
15
  "vision_config": {
16
  "hidden_size": 168,
17
  "mlp_dim": 696,
 
11
  "prompt_encoder_config": {
12
  "model_type": ""
13
  },
14
+ "transformers_version": "4.37.2",
15
  "vision_config": {
16
  "hidden_size": 168,
17
  "mlp_dim": 696,
onnx/prompt_encoder_mask_decoder.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f4514391764fbd56e08e119060d874ecd7d52994bfb1968af159e12d4943b5bb
3
- size 16557892
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b8830426d85bc4b4e49795203be8915c291bd78edfdb9c34285d7c1f7d3d0487
3
+ size 16595228
onnx/prompt_encoder_mask_decoder_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:cb90b279f549d2cab7fd6e20c38522438c65d84bdcca3d2a764cff7d857fdce2
3
- size 4903810
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6f6a3df6b607572bf9eab9f0ca77e9c79fef67ba308306e050766fe996a54aa9
3
+ size 4956135
onnx/vision_encoder.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:9f8433273a6750b587779baa0cf5508111001bf7e7acfcf585d370139fd366d0
3
- size 23276014
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:24bd98fd8e375b257aed9550ef2b25bf026004bdf1e233bfc15d0475c627441f
3
+ size 23275998
onnx/vision_encoder_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:cce23c7b2e5d4f330932738fb67ba518e04b0d99ccdd1cccd22a7da4e01f2971
3
- size 8882165
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0a1cbc7db99bec8baa85ec2368c72230e75dc17b664ecec539e6fbbd8aa1ab28
3
+ size 8882149
preprocessor_config.json CHANGED
@@ -15,6 +15,13 @@
15
  0.224,
16
  0.225
17
  ],
 
 
 
 
 
 
 
18
  "pad_size": {
19
  "height": 1024,
20
  "width": 1024
 
15
  0.224,
16
  0.225
17
  ],
18
+ "mask_pad_size": {
19
+ "height": 256,
20
+ "width": 256
21
+ },
22
+ "mask_size": {
23
+ "longest_edge": 256
24
+ },
25
  "pad_size": {
26
  "height": 1024,
27
  "width": 1024
quantize_config.json CHANGED
@@ -2,81 +2,79 @@
2
  "per_channel": true,
3
  "reduce_range": true,
4
  "per_model_config": {
5
- "prompt_encoder_mask_decoder": {
6
  "op_types": [
7
- "Gather",
8
- "Sin",
9
- "Unsqueeze",
10
- "Not",
11
- "Cast",
12
- "Equal",
13
- "Neg",
14
- "MatMul",
15
- "Cos",
16
- "OneHot",
17
  "Transpose",
18
- "Sub",
19
- "Where",
20
- "Erf",
21
  "ConstantOfShape",
22
- "ReduceMean",
23
- "Shape",
24
- "Range",
 
 
25
  "Expand",
26
- "Div",
27
- "Relu",
28
- "ConvTranspose",
 
 
 
29
  "Tile",
30
- "Concat",
31
- "ScatterND",
32
- "Sqrt",
33
  "Slice",
 
34
  "Mul",
35
- "Constant",
36
- "Pow",
37
- "Reshape",
 
 
 
 
 
 
 
38
  "Add",
39
- "Softmax"
40
  ],
41
- "weight_type": "QInt8"
42
  },
43
- "vision_encoder": {
44
  "op_types": [
45
- "Gather",
46
- "Sin",
47
- "Unsqueeze",
48
- "Cast",
49
- "MatMul",
50
- "Cos",
51
  "Transpose",
52
- "Sub",
53
- "Pad",
54
- "Erf",
55
- "Conv",
56
  "ConstantOfShape",
57
- "Resize",
58
- "ReduceMean",
59
- "Shape",
60
- "Range",
61
- "Expand",
62
- "Div",
63
- "Tile",
64
- "Mod",
65
- "Squeeze",
66
  "Concat",
67
- "Sqrt",
68
- "Split",
69
- "Mul",
70
- "Slice",
71
- "Identity",
72
  "Constant",
 
73
  "Pow",
74
- "Einsum",
75
- "Reshape",
 
 
 
 
 
 
 
 
 
 
 
76
  "Add",
77
- "Softmax"
 
 
78
  ],
79
- "weight_type": "QUInt8"
80
  }
81
  }
82
  }
 
2
  "per_channel": true,
3
  "reduce_range": true,
4
  "per_model_config": {
5
+ "vision_encoder": {
6
  "op_types": [
7
+ "Div",
8
+ "Softmax",
 
 
 
 
 
 
 
 
9
  "Transpose",
 
 
 
10
  "ConstantOfShape",
11
+ "Erf",
12
+ "Concat",
13
+ "Identity",
14
+ "Reshape",
15
+ "Pad",
16
  "Expand",
17
+ "ReduceMean",
18
+ "Unsqueeze",
19
+ "Conv",
20
+ "Resize",
21
+ "Constant",
22
+ "Pow",
23
  "Tile",
24
+ "Split",
 
 
25
  "Slice",
26
+ "Range",
27
  "Mul",
28
+ "Sub",
29
+ "Sqrt",
30
+ "Cos",
31
+ "Sin",
32
+ "Mod",
33
+ "Gather",
34
+ "Shape",
35
+ "Squeeze",
36
+ "Cast",
37
+ "Einsum",
38
  "Add",
39
+ "MatMul"
40
  ],
41
+ "weight_type": "QUInt8"
42
  },
43
+ "prompt_encoder_mask_decoder": {
44
  "op_types": [
45
+ "Div",
46
+ "Softmax",
 
 
 
 
47
  "Transpose",
48
+ "ConvTranspose",
 
 
 
49
  "ConstantOfShape",
50
+ "Erf",
 
 
 
 
 
 
 
 
51
  "Concat",
52
+ "Reshape",
53
+ "Expand",
54
+ "ReduceMean",
55
+ "Unsqueeze",
 
56
  "Constant",
57
+ "Relu",
58
  "Pow",
59
+ "Tile",
60
+ "Slice",
61
+ "Range",
62
+ "Mul",
63
+ "Sub",
64
+ "Equal",
65
+ "Sqrt",
66
+ "Sin",
67
+ "Cos",
68
+ "Gather",
69
+ "OneHot",
70
+ "Shape",
71
+ "Cast",
72
  "Add",
73
+ "MatMul",
74
+ "ScatterND",
75
+ "Where"
76
  ],
77
+ "weight_type": "QInt8"
78
  }
79
  }
80
  }