Hugging Face's logo

hf-tiny-v2
/
tiny-random-Sam3TrackerModel

Safetensors
sam3_tracker
Model card Files
xet
 Community
hf-transformers-bot HF Staff commited on
Commit
efca257
·
verified ·
1 Parent(s): fc6e498

Update tiny models for Sam3TrackerModel

Browse files
Files changed (6) hide show
  1. config.json +108 -0
  2. model.safetensors +3 -0
  3. preprocessor_config.json +31 -0
  4. tokenizer.json +0 -0
  5. tokenizer_config.json +17 -0
  6. video_preprocessor_config.json +28 -0
config.json ADDED
@@ -0,0 +1,108 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Sam3TrackerModel"
4
+ ],
5
+ "dtype": "float32",
6
+ "image_size": 224,
7
+ "initializer_range": 0.02,
8
+ "mask_decoder_config": {
9
+ "attention_downsample_rate": 2,
10
+ "dynamic_multimask_stability_delta": 0.05,
11
+ "dynamic_multimask_stability_thresh": 0.98,
12
+ "dynamic_multimask_via_stability": true,
13
+ "hidden_act": "relu",
14
+ "hidden_size": 32,
15
+ "iou_head_depth": 3,
16
+ "iou_head_hidden_dim": 32,
17
+ "mlp_dim": 64,
18
+ "model_type": "",
19
+ "num_attention_heads": 4,
20
+ "num_hidden_layers": 2,
21
+ "num_multimask_outputs": 3
22
+ },
23
+ "mask_downsampler_embed_dim": 32,
24
+ "memory_attention_feed_forward_hidden_size": 32,
25
+ "memory_attention_hidden_size": 32,
26
+ "memory_attention_num_layers": 1,
27
+ "memory_encoder_hidden_size": 32,
28
+ "memory_fuser_embed_dim": 32,
29
+ "model_type": "sam3_tracker",
30
+ "prompt_encoder_config": {
31
+ "hidden_act": "gelu",
32
+ "hidden_size": 32,
33
+ "image_size": 128,
34
+ "layer_norm_eps": 1e-06,
35
+ "mask_input_channels": 8,
36
+ "model_type": "",
37
+ "num_point_embeddings": 4,
38
+ "patch_size": 16,
39
+ "scale": 1
40
+ },
41
+ "transformers_version": "5.16.0.dev0",
42
+ "vision_config": {
43
+ "backbone_config": {
44
+ "_name_or_path": "",
45
+ "architectures": null,
46
+ "attention_dropout": 0.0,
47
+ "chunk_size_feed_forward": 0,
48
+ "dtype": null,
49
+ "global_attn_indexes": [
50
+ 0,
51
+ 1
52
+ ],
53
+ "hidden_act": "gelu",
54
+ "hidden_dropout": 0.0,
55
+ "hidden_size": 32,
56
+ "id2label": {
57
+ "0": "LABEL_0",
58
+ "1": "LABEL_1"
59
+ },
60
+ "image_size": 224,
61
+ "initializer_range": 0.02,
62
+ "intermediate_size": 64,
63
+ "is_encoder_decoder": false,
64
+ "label2id": {
65
+ "LABEL_0": 0,
66
+ "LABEL_1": 1
67
+ },
68
+ "layer_norm_eps": 1e-06,
69
+ "layer_scale_init_value": null,
70
+ "model_type": "sam3_vit_model",
71
+ "num_attention_heads": 4,
72
+ "num_channels": 3,
73
+ "num_hidden_layers": 2,
74
+ "output_attentions": false,
75
+ "output_hidden_states": false,
76
+ "patch_size": 14,
77
+ "pretrain_image_size": 336,
78
+ "problem_type": null,
79
+ "return_dict": true,
80
+ "rope_theta": 10000.0,
81
+ "window_size": 8
82
+ },
83
+ "backbone_feature_sizes": [
84
+ [
85
+ 32,
86
+ 32
87
+ ],
88
+ [
89
+ 16,
90
+ 16
91
+ ],
92
+ [
93
+ 8,
94
+ 8
95
+ ]
96
+ ],
97
+ "fpn_hidden_size": 32,
98
+ "hidden_act": "gelu",
99
+ "initializer_range": 0.02,
100
+ "layer_norm_eps": 1e-06,
101
+ "model_type": "sam3_vision_model",
102
+ "scale_factors": [
103
+ 2.0,
104
+ 1.0,
105
+ 0.5
106
+ ]
107
+ }
108
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9725d5eac97bfb495c949e7f3f2cfe9a31aa35bbf357de9b95432f8156e63a4a
3
+ size 549236
preprocessor_config.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "crop_size": {
3
+ "height": 224,
4
+ "width": 224
5
+ },
6
+ "do_convert_rgb": true,
7
+ "do_normalize": true,
8
+ "do_rescale": true,
9
+ "do_resize": true,
10
+ "image_mean": [
11
+ 0.5,
12
+ 0.5,
13
+ 0.5
14
+ ],
15
+ "image_processor_type": "Sam3ImageProcessor",
16
+ "image_std": [
17
+ 0.5,
18
+ 0.5,
19
+ 0.5
20
+ ],
21
+ "mask_size": {
22
+ "height": 288,
23
+ "width": 288
24
+ },
25
+ "resample": 2,
26
+ "rescale_factor": 0.00392156862745098,
27
+ "size": {
28
+ "height": 224,
29
+ "width": 224
30
+ }
31
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,17 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "backend": "tokenizers",
4
+ "bos_token": "<|startoftext|>",
5
+ "clean_up_tokenization_spaces": false,
6
+ "do_lower_case": true,
7
+ "eos_token": "<|endoftext|>",
8
+ "errors": "replace",
9
+ "is_local": true,
10
+ "local_files_only": false,
11
+ "max_length": 32,
12
+ "model_max_length": 32,
13
+ "pad_token": "<|endoftext|>",
14
+ "processor_class": "Sam3VideoProcessor",
15
+ "tokenizer_class": "CLIPTokenizer",
16
+ "unk_token": "<|endoftext|>"
17
+ }
video_preprocessor_config.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "crop_size": {
3
+ "height": 224,
4
+ "width": 224
5
+ },
6
+ "do_convert_rgb": true,
7
+ "do_normalize": true,
8
+ "do_rescale": true,
9
+ "do_resize": true,
10
+ "image_mean": [
11
+ 0.485,
12
+ 0.456,
13
+ 0.406
14
+ ],
15
+ "image_std": [
16
+ 0.229,
17
+ 0.224,
18
+ 0.225
19
+ ],
20
+ "resample": 2,
21
+ "rescale_factor": 0.00392156862745098,
22
+ "return_metadata": false,
23
+ "size": {
24
+ "height": 224,
25
+ "width": 224
26
+ },
27
+ "video_processor_type": "Sam2VideoVideoProcessor"
28
+ }