Hugging Face's logo

facebook
/
sapiens2-normal-0.4b

Depth Estimation
sapiens2
Safetensors
sapiens
Transformers
human-centric
normal
Model card Files
xet
 Community
4

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

#1
by guarin HF Staff - opened
base: refs/heads/main
←
from: refs/pr/1
Discussion Files changed
+196
-0
Files changed (3) hide show
  1. config.json +164 -0
  2. model.safetensors +3 -0
  3. preprocessor_config.json +29 -0
config.json ADDED
@@ -0,0 +1,164 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "apply_layernorm": true,
3
+ "architectures": [
4
+ "Sapiens2ForNormalEstimation"
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": 1024,
61
+ "id2label": {
62
+ "0": "LABEL_0",
63
+ "1": "LABEL_1",
64
+ "2": "LABEL_2"
65
+ },
66
+ "image_size": [
67
+ 1024,
68
+ 768
69
+ ],
70
+ "initializer_range": 0.02,
71
+ "intermediate_size": 4096,
72
+ "key_bias": true,
73
+ "label2id": {
74
+ "LABEL_0": 0,
75
+ "LABEL_1": 1,
76
+ "LABEL_2": 2
77
+ },
78
+ "layer_norm_eps": 1e-05,
79
+ "layerscale_value": 1.0,
80
+ "mlp_bias": true,
81
+ "model_type": "sapiens2",
82
+ "normalize_backbone_outputs": true,
83
+ "num_attention_heads": 16,
84
+ "num_channels": 3,
85
+ "num_first_full_attention_layers": 8,
86
+ "num_hidden_layers": 24,
87
+ "num_key_value_attention_heads": 8,
88
+ "num_key_value_heads_per_layer": [
89
+ 16,
90
+ 16,
91
+ 16,
92
+ 16,
93
+ 16,
94
+ 16,
95
+ 16,
96
+ 16,
97
+ 8,
98
+ 8,
99
+ 8,
100
+ 8,
101
+ 8,
102
+ 8,
103
+ 8,
104
+ 8,
105
+ 16,
106
+ 16,
107
+ 16,
108
+ 16,
109
+ 16,
110
+ 16,
111
+ 16,
112
+ 16
113
+ ],
114
+ "num_last_full_attention_layers": 8,
115
+ "num_register_tokens": 8,
116
+ "out_features": [
117
+ "stage24"
118
+ ],
119
+ "out_indices": [
120
+ 24
121
+ ],
122
+ "patch_size": 16,
123
+ "pos_embed_jitter": null,
124
+ "pos_embed_rescale": 2.0,
125
+ "pos_embed_shift": null,
126
+ "proj_bias": true,
127
+ "query_bias": true,
128
+ "reshape_hidden_states": true,
129
+ "rms_norm_eps": 1e-06,
130
+ "rope_theta": 100.0,
131
+ "semantic_loss_ignore_index": 255,
132
+ "stage_names": [
133
+ "stem",
134
+ "stage1",
135
+ "stage2",
136
+ "stage3",
137
+ "stage4",
138
+ "stage5",
139
+ "stage6",
140
+ "stage7",
141
+ "stage8",
142
+ "stage9",
143
+ "stage10",
144
+ "stage11",
145
+ "stage12",
146
+ "stage13",
147
+ "stage14",
148
+ "stage15",
149
+ "stage16",
150
+ "stage17",
151
+ "stage18",
152
+ "stage19",
153
+ "stage20",
154
+ "stage21",
155
+ "stage22",
156
+ "stage23",
157
+ "stage24"
158
+ ],
159
+ "transformers_version": "5.10.0.dev0",
160
+ "use_gated_mlp": true,
161
+ "use_mask_token": false,
162
+ "use_qk_norm": true,
163
+ "value_bias": true
164
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d2f20e8df7186772b7710546416db0752be224c68003c7f18ccbbf8b980ef1fd
3
+ size 1813476476
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": true,
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
+ }