Hugging Face's logo

facebook
/
sapiens2-matting-1b

Image Segmentation
sapiens2
Safetensors
sapiens
Transformers
human-centric
matting
Model card Files
xet
 Community
4
rawalkhirodkar guarin HF Staff commited on
Commit
c812f41
·
1 Parent(s): 1b729ef

Add HF-compatible model.safetensors, config.json, and preprocessor_config.json (#1)

Browse files

- Add HF-compatible model.safetensors, config.json, and preprocessor_config.json (0ba513a9f8521de749a6155dc33b246e4b32c42b)
- Remove 'pos_embed_dtype' from config.json (d66e70fb1055a1bd2b6b154353a1aa319b9cf06a)
- Update config.json and preprocessor_config.json (ac734336ce3c28189d399ecc50fe461ad48134da)
- Update config.json and preprocessor_config.json (61a55f1c78042911540b5afb5f768ea941075079)
- Update config.json and preprocessor_config.json (1acc3439fc04b45824c95c5662e3c8780b837ca4)
- Update config.json and preprocessor_config.json (9699ad600d4db998e2df7e3d79e2a4904de76bfa)


Co-authored-by: guarin <guarin@users.noreply.huggingface.co>

Files changed (3) hide show
  1. config.json +198 -0
  2. model.safetensors +3 -0
  3. preprocessor_config.json +29 -0
config.json ADDED
@@ -0,0 +1,198 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "apply_layernorm": true,
3
+ "architectures": [
4
+ "Sapiens2ForImageMatting"
5
+ ],
6
+ "attention_dropout": 0.0,
7
+ "drop_path_rate": 0.0,
8
+ "flip_pairs": null,
9
+ "head_config": {
10
+ "_name_or_path": "",
11
+ "architectures": null,
12
+ "chunk_size_feed_forward": 0,
13
+ "conv_kernel_size": 1,
14
+ "conv_kernel_sizes": [
15
+ 3,
16
+ 3,
17
+ 3
18
+ ],
19
+ "conv_out_channels": [
20
+ 64,
21
+ 32,
22
+ 16
23
+ ],
24
+ "dtype": null,
25
+ "id2label": {
26
+ "0": "LABEL_0",
27
+ "1": "LABEL_1"
28
+ },
29
+ "is_encoder_decoder": false,
30
+ "label2id": {
31
+ "LABEL_0": 0,
32
+ "LABEL_1": 1
33
+ },
34
+ "model_type": "sapiens2_head",
35
+ "output_attentions": false,
36
+ "output_hidden_states": false,
37
+ "problem_type": null,
38
+ "return_dict": true,
39
+ "scale_conv_kernel_size": 1,
40
+ "scale_conv_kernel_sizes": null,
41
+ "scale_conv_out_channels": null,
42
+ "scale_final_hidden_sizes": null,
43
+ "scale_final_input_size": null,
44
+ "upsample_kernel_size": 4,
45
+ "upsample_kernel_sizes": [
46
+ 3,
47
+ 3,
48
+ 3,
49
+ 3
50
+ ],
51
+ "upsample_out_channels": [
52
+ 768,
53
+ 512,
54
+ 256,
55
+ 128
56
+ ],
57
+ "use_pixel_shuffle": true
58
+ },
59
+ "hidden_act": "silu",
60
+ "hidden_size": 1536,
61
+ "id2label": {
62
+ "0": "LABEL_0",
63
+ "1": "LABEL_1",
64
+ "2": "LABEL_2",
65
+ "3": "LABEL_3"
66
+ },
67
+ "image_size": [
68
+ 1024,
69
+ 768
70
+ ],
71
+ "initializer_range": 0.02,
72
+ "intermediate_size": 6144,
73
+ "key_bias": true,
74
+ "label2id": {
75
+ "LABEL_0": 0,
76
+ "LABEL_1": 1,
77
+ "LABEL_2": 2,
78
+ "LABEL_3": 3
79
+ },
80
+ "layer_norm_eps": 1e-05,
81
+ "layerscale_value": 1.0,
82
+ "mlp_bias": true,
83
+ "model_type": "sapiens2",
84
+ "normalize_backbone_outputs": true,
85
+ "num_attention_heads": 24,
86
+ "num_channels": 3,
87
+ "num_first_full_attention_layers": 8,
88
+ "num_hidden_layers": 40,
89
+ "num_key_value_attention_heads": 12,
90
+ "num_key_value_heads_per_layer": [
91
+ 24,
92
+ 24,
93
+ 24,
94
+ 24,
95
+ 24,
96
+ 24,
97
+ 24,
98
+ 24,
99
+ 12,
100
+ 12,
101
+ 12,
102
+ 12,
103
+ 12,
104
+ 12,
105
+ 12,
106
+ 12,
107
+ 12,
108
+ 12,
109
+ 12,
110
+ 12,
111
+ 12,
112
+ 12,
113
+ 12,
114
+ 12,
115
+ 12,
116
+ 12,
117
+ 12,
118
+ 12,
119
+ 12,
120
+ 12,
121
+ 12,
122
+ 12,
123
+ 24,
124
+ 24,
125
+ 24,
126
+ 24,
127
+ 24,
128
+ 24,
129
+ 24,
130
+ 24
131
+ ],
132
+ "num_last_full_attention_layers": 8,
133
+ "num_register_tokens": 8,
134
+ "out_features": [
135
+ "stage40"
136
+ ],
137
+ "out_indices": [
138
+ 40
139
+ ],
140
+ "patch_size": 16,
141
+ "pos_embed_jitter": null,
142
+ "pos_embed_rescale": 2.0,
143
+ "pos_embed_shift": null,
144
+ "proj_bias": true,
145
+ "query_bias": true,
146
+ "reshape_hidden_states": true,
147
+ "rms_norm_eps": 1e-06,
148
+ "rope_theta": 100.0,
149
+ "semantic_loss_ignore_index": 255,
150
+ "stage_names": [
151
+ "stem",
152
+ "stage1",
153
+ "stage2",
154
+ "stage3",
155
+ "stage4",
156
+ "stage5",
157
+ "stage6",
158
+ "stage7",
159
+ "stage8",
160
+ "stage9",
161
+ "stage10",
162
+ "stage11",
163
+ "stage12",
164
+ "stage13",
165
+ "stage14",
166
+ "stage15",
167
+ "stage16",
168
+ "stage17",
169
+ "stage18",
170
+ "stage19",
171
+ "stage20",
172
+ "stage21",
173
+ "stage22",
174
+ "stage23",
175
+ "stage24",
176
+ "stage25",
177
+ "stage26",
178
+ "stage27",
179
+ "stage28",
180
+ "stage29",
181
+ "stage30",
182
+ "stage31",
183
+ "stage32",
184
+ "stage33",
185
+ "stage34",
186
+ "stage35",
187
+ "stage36",
188
+ "stage37",
189
+ "stage38",
190
+ "stage39",
191
+ "stage40"
192
+ ],
193
+ "transformers_version": "5.10.0.dev0",
194
+ "use_gated_mlp": true,
195
+ "use_mask_token": false,
196
+ "use_qk_norm": true,
197
+ "value_bias": true
198
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2a0ea4ea2234e8dbd51e952950d8cedfe3ee2eca9099003784fd2d9cec6bc6e1
3
+ size 6157412312
preprocessor_config.json ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "crop_size": {
3
+ "height": 224,
4
+ "width": 224
5
+ },
6
+ "do_center_crop": false,
7
+ "do_normalize": true,
8
+ "do_pad": false,
9
+ "do_reduce_labels": false,
10
+ "do_rescale": true,
11
+ "do_resize": true,
12
+ "image_mean": [
13
+ 0.485,
14
+ 0.456,
15
+ 0.406
16
+ ],
17
+ "image_processor_type": "Sapiens2ImageProcessor",
18
+ "image_std": [
19
+ 0.229,
20
+ 0.224,
21
+ 0.225
22
+ ],
23
+ "resample": 2,
24
+ "rescale_factor": 0.00392156862745098,
25
+ "size": {
26
+ "height": 1024,
27
+ "width": 768
28
+ }
29
+ }