Hugging Face's logo

facebook
/
sapiens2-pretrain-0.8b

Image Feature Extraction
sapiens2
Safetensors
sapiens
Transformers
feature-extraction
vision-transformer
human-centric
pretrained-backbone
Model card Files
xet
 Community
4
rawalkhirodkar guarin HF Staff commited on
Commit
fd1847f
·
1 Parent(s): bca95d5

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 (79ffb93cbb7df21dc438ae65d44a17eee5572b94)
- Remove 'pos_embed_dtype' from config.json (36bdd797f88b943ae172c605d54b81e902c5d488)
- Update config.json and preprocessor_config.json (bf434fec1bd22de78ff7c4a709dcb5a290dcb492)
- Update config.json and preprocessor_config.json (9120dd1a05a7f30756501b7af274a0b42b95bd9d)


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

Files changed (3) hide show
  1. config.json +121 -0
  2. model.safetensors +3 -0
  3. preprocessor_config.json +29 -0
config.json ADDED
@@ -0,0 +1,121 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "apply_layernorm": true,
3
+ "architectures": [
4
+ "Sapiens2Model"
5
+ ],
6
+ "attention_dropout": 0.0,
7
+ "drop_path_rate": 0.0,
8
+ "flip_pairs": null,
9
+ "head_config": null,
10
+ "hidden_act": "silu",
11
+ "hidden_size": 1280,
12
+ "image_size": [
13
+ 1024,
14
+ 768
15
+ ],
16
+ "initializer_range": 0.02,
17
+ "intermediate_size": 5120,
18
+ "key_bias": true,
19
+ "layer_norm_eps": 1e-05,
20
+ "layerscale_value": 1.0,
21
+ "mlp_bias": true,
22
+ "model_type": "sapiens2",
23
+ "normalize_backbone_outputs": true,
24
+ "num_attention_heads": 16,
25
+ "num_channels": 3,
26
+ "num_first_full_attention_layers": 8,
27
+ "num_hidden_layers": 32,
28
+ "num_key_value_attention_heads": 8,
29
+ "num_key_value_heads_per_layer": [
30
+ 16,
31
+ 16,
32
+ 16,
33
+ 16,
34
+ 16,
35
+ 16,
36
+ 16,
37
+ 16,
38
+ 8,
39
+ 8,
40
+ 8,
41
+ 8,
42
+ 8,
43
+ 8,
44
+ 8,
45
+ 8,
46
+ 8,
47
+ 8,
48
+ 8,
49
+ 8,
50
+ 8,
51
+ 8,
52
+ 8,
53
+ 8,
54
+ 16,
55
+ 16,
56
+ 16,
57
+ 16,
58
+ 16,
59
+ 16,
60
+ 16,
61
+ 16
62
+ ],
63
+ "num_last_full_attention_layers": 8,
64
+ "num_register_tokens": 8,
65
+ "out_features": [
66
+ "stage32"
67
+ ],
68
+ "out_indices": [
69
+ 32
70
+ ],
71
+ "patch_size": 16,
72
+ "pos_embed_jitter": null,
73
+ "pos_embed_rescale": 2.0,
74
+ "pos_embed_shift": null,
75
+ "proj_bias": true,
76
+ "query_bias": true,
77
+ "reshape_hidden_states": true,
78
+ "rms_norm_eps": 1e-06,
79
+ "rope_theta": 100.0,
80
+ "semantic_loss_ignore_index": 255,
81
+ "stage_names": [
82
+ "stem",
83
+ "stage1",
84
+ "stage2",
85
+ "stage3",
86
+ "stage4",
87
+ "stage5",
88
+ "stage6",
89
+ "stage7",
90
+ "stage8",
91
+ "stage9",
92
+ "stage10",
93
+ "stage11",
94
+ "stage12",
95
+ "stage13",
96
+ "stage14",
97
+ "stage15",
98
+ "stage16",
99
+ "stage17",
100
+ "stage18",
101
+ "stage19",
102
+ "stage20",
103
+ "stage21",
104
+ "stage22",
105
+ "stage23",
106
+ "stage24",
107
+ "stage25",
108
+ "stage26",
109
+ "stage27",
110
+ "stage28",
111
+ "stage29",
112
+ "stage30",
113
+ "stage31",
114
+ "stage32"
115
+ ],
116
+ "transformers_version": "5.10.0.dev0",
117
+ "use_gated_mlp": true,
118
+ "use_mask_token": false,
119
+ "use_qk_norm": true,
120
+ "value_bias": true
121
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a628d3e56aec159ceb95892bc6a594bd890fd703b2184b49dd404566286ea8cc
3
+ size 3257187528
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
+ }