Add HF-compatible model.safetensors, config.json, and preprocessor_config.json

#1
by guarin HF Staff - opened
Files changed (3) hide show
  1. config.json +246 -0
  2. model.safetensors +3 -0
  3. preprocessor_config.json +29 -0
config.json ADDED
@@ -0,0 +1,246 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "apply_layernorm": true,
3
+ "architectures": [
4
+ "Sapiens2ForSemanticSegmentation"
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
+ 1,
16
+ 1
17
+ ],
18
+ "conv_out_channels": [
19
+ 64,
20
+ 64
21
+ ],
22
+ "dtype": null,
23
+ "id2label": {
24
+ "0": "LABEL_0",
25
+ "1": "LABEL_1"
26
+ },
27
+ "is_encoder_decoder": false,
28
+ "label2id": {
29
+ "LABEL_0": 0,
30
+ "LABEL_1": 1
31
+ },
32
+ "model_type": "sapiens2_head",
33
+ "output_attentions": false,
34
+ "output_hidden_states": false,
35
+ "problem_type": null,
36
+ "return_dict": true,
37
+ "scale_conv_kernel_size": 1,
38
+ "scale_conv_kernel_sizes": null,
39
+ "scale_conv_out_channels": null,
40
+ "scale_final_hidden_sizes": null,
41
+ "scale_final_input_size": null,
42
+ "upsample_kernel_size": 4,
43
+ "upsample_kernel_sizes": [
44
+ 4,
45
+ 4,
46
+ 4,
47
+ 4
48
+ ],
49
+ "upsample_out_channels": [
50
+ 512,
51
+ 256,
52
+ 128,
53
+ 64
54
+ ],
55
+ "use_pixel_shuffle": null
56
+ },
57
+ "hidden_act": "silu",
58
+ "hidden_size": 1536,
59
+ "id2label": {
60
+ "0": "LABEL_0",
61
+ "1": "LABEL_1",
62
+ "2": "LABEL_2",
63
+ "3": "LABEL_3",
64
+ "4": "LABEL_4",
65
+ "5": "LABEL_5",
66
+ "6": "LABEL_6",
67
+ "7": "LABEL_7",
68
+ "8": "LABEL_8",
69
+ "9": "LABEL_9",
70
+ "10": "LABEL_10",
71
+ "11": "LABEL_11",
72
+ "12": "LABEL_12",
73
+ "13": "LABEL_13",
74
+ "14": "LABEL_14",
75
+ "15": "LABEL_15",
76
+ "16": "LABEL_16",
77
+ "17": "LABEL_17",
78
+ "18": "LABEL_18",
79
+ "19": "LABEL_19",
80
+ "20": "LABEL_20",
81
+ "21": "LABEL_21",
82
+ "22": "LABEL_22",
83
+ "23": "LABEL_23",
84
+ "24": "LABEL_24",
85
+ "25": "LABEL_25",
86
+ "26": "LABEL_26",
87
+ "27": "LABEL_27",
88
+ "28": "LABEL_28"
89
+ },
90
+ "image_size": [
91
+ 1024,
92
+ 768
93
+ ],
94
+ "initializer_range": 0.02,
95
+ "intermediate_size": 6144,
96
+ "key_bias": true,
97
+ "label2id": {
98
+ "LABEL_0": 0,
99
+ "LABEL_1": 1,
100
+ "LABEL_10": 10,
101
+ "LABEL_11": 11,
102
+ "LABEL_12": 12,
103
+ "LABEL_13": 13,
104
+ "LABEL_14": 14,
105
+ "LABEL_15": 15,
106
+ "LABEL_16": 16,
107
+ "LABEL_17": 17,
108
+ "LABEL_18": 18,
109
+ "LABEL_19": 19,
110
+ "LABEL_2": 2,
111
+ "LABEL_20": 20,
112
+ "LABEL_21": 21,
113
+ "LABEL_22": 22,
114
+ "LABEL_23": 23,
115
+ "LABEL_24": 24,
116
+ "LABEL_25": 25,
117
+ "LABEL_26": 26,
118
+ "LABEL_27": 27,
119
+ "LABEL_28": 28,
120
+ "LABEL_3": 3,
121
+ "LABEL_4": 4,
122
+ "LABEL_5": 5,
123
+ "LABEL_6": 6,
124
+ "LABEL_7": 7,
125
+ "LABEL_8": 8,
126
+ "LABEL_9": 9
127
+ },
128
+ "layer_norm_eps": 1e-05,
129
+ "layerscale_value": 1.0,
130
+ "mlp_bias": true,
131
+ "model_type": "sapiens2",
132
+ "normalize_backbone_outputs": true,
133
+ "num_attention_heads": 24,
134
+ "num_channels": 3,
135
+ "num_first_full_attention_layers": 8,
136
+ "num_hidden_layers": 40,
137
+ "num_key_value_attention_heads": 12,
138
+ "num_key_value_heads_per_layer": [
139
+ 24,
140
+ 24,
141
+ 24,
142
+ 24,
143
+ 24,
144
+ 24,
145
+ 24,
146
+ 24,
147
+ 12,
148
+ 12,
149
+ 12,
150
+ 12,
151
+ 12,
152
+ 12,
153
+ 12,
154
+ 12,
155
+ 12,
156
+ 12,
157
+ 12,
158
+ 12,
159
+ 12,
160
+ 12,
161
+ 12,
162
+ 12,
163
+ 12,
164
+ 12,
165
+ 12,
166
+ 12,
167
+ 12,
168
+ 12,
169
+ 12,
170
+ 12,
171
+ 24,
172
+ 24,
173
+ 24,
174
+ 24,
175
+ 24,
176
+ 24,
177
+ 24,
178
+ 24
179
+ ],
180
+ "num_last_full_attention_layers": 8,
181
+ "num_register_tokens": 8,
182
+ "out_features": [
183
+ "stage40"
184
+ ],
185
+ "out_indices": [
186
+ 40
187
+ ],
188
+ "patch_size": 16,
189
+ "pos_embed_jitter": null,
190
+ "pos_embed_rescale": 2.0,
191
+ "pos_embed_shift": null,
192
+ "proj_bias": true,
193
+ "query_bias": true,
194
+ "reshape_hidden_states": true,
195
+ "rms_norm_eps": 1e-06,
196
+ "rope_theta": 100.0,
197
+ "semantic_loss_ignore_index": 255,
198
+ "stage_names": [
199
+ "stem",
200
+ "stage1",
201
+ "stage2",
202
+ "stage3",
203
+ "stage4",
204
+ "stage5",
205
+ "stage6",
206
+ "stage7",
207
+ "stage8",
208
+ "stage9",
209
+ "stage10",
210
+ "stage11",
211
+ "stage12",
212
+ "stage13",
213
+ "stage14",
214
+ "stage15",
215
+ "stage16",
216
+ "stage17",
217
+ "stage18",
218
+ "stage19",
219
+ "stage20",
220
+ "stage21",
221
+ "stage22",
222
+ "stage23",
223
+ "stage24",
224
+ "stage25",
225
+ "stage26",
226
+ "stage27",
227
+ "stage28",
228
+ "stage29",
229
+ "stage30",
230
+ "stage31",
231
+ "stage32",
232
+ "stage33",
233
+ "stage34",
234
+ "stage35",
235
+ "stage36",
236
+ "stage37",
237
+ "stage38",
238
+ "stage39",
239
+ "stage40"
240
+ ],
241
+ "transformers_version": "5.10.0.dev0",
242
+ "use_gated_mlp": true,
243
+ "use_mask_token": false,
244
+ "use_qk_norm": true,
245
+ "value_bias": true
246
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4b73c44963b377e93fcb4c4053f72a189836a22d05e12c30383046b9cd3c5bd4
3
+ size 5883353380
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
+ }