Hugging Face's logo

hf-tiny-v2
/
tiny-random-Sam3LiteTextModel

Safetensors
sam3_lite_text
Model card Files
xet
 Community
hf-transformers-bot commited on
Commit
891074c
·
verified ·
1 Parent(s): 424b662

Update tiny models for Sam3LiteTextModel

Browse files
Files changed (5) hide show
  1. config.json +139 -0
  2. model.safetensors +3 -0
  3. preprocessor_config.json +31 -0
  4. tokenizer.json +0 -0
  5. tokenizer_config.json +16 -0
config.json ADDED
@@ -0,0 +1,139 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Sam3LiteTextModel"
4
+ ],
5
+ "detr_decoder_config": {
6
+ "dropout": 0.1,
7
+ "hidden_act": "relu",
8
+ "hidden_dropout": 0.0,
9
+ "hidden_size": 32,
10
+ "initializer_range": 0.02,
11
+ "intermediate_size": 64,
12
+ "layer_norm_eps": 1e-06,
13
+ "model_type": "sam3_lite_text_detr_decoder",
14
+ "num_attention_heads": 4,
15
+ "num_layers": 1,
16
+ "num_queries": 5
17
+ },
18
+ "detr_encoder_config": {
19
+ "dropout": 0.1,
20
+ "hidden_act": "relu",
21
+ "hidden_dropout": 0.0,
22
+ "hidden_size": 32,
23
+ "initializer_range": 0.02,
24
+ "intermediate_size": 64,
25
+ "layer_norm_eps": 1e-06,
26
+ "model_type": "sam3_lite_text_detr_encoder",
27
+ "num_attention_heads": 4,
28
+ "num_layers": 1
29
+ },
30
+ "dtype": "float32",
31
+ "geometry_encoder_config": {
32
+ "dropout": 0.1,
33
+ "hidden_act": "relu",
34
+ "hidden_dropout": 0.0,
35
+ "hidden_size": 32,
36
+ "initializer_range": 0.02,
37
+ "intermediate_size": 64,
38
+ "layer_norm_eps": 1e-06,
39
+ "mask_fuser_hidden_size": 32,
40
+ "mask_fuser_num_layers": 1,
41
+ "model_type": "sam3_lite_text_geometry_encoder",
42
+ "num_attention_heads": 4,
43
+ "num_layers": 1,
44
+ "roi_size": 7
45
+ },
46
+ "initializer_range": 0.02,
47
+ "mask_decoder_config": {
48
+ "dropout": 0.0,
49
+ "hidden_size": 32,
50
+ "initializer_range": 0.02,
51
+ "layer_norm_eps": 1e-06,
52
+ "model_type": "sam3_lite_text_mask_decoder",
53
+ "num_attention_heads": 8,
54
+ "num_upsampling_stages": 2
55
+ },
56
+ "model_type": "sam3_lite_text",
57
+ "text_config": {
58
+ "attention_dropout": 0.0,
59
+ "hidden_act": "gelu",
60
+ "hidden_size": 32,
61
+ "intermediate_size": 64,
62
+ "layer_norm_eps": 1e-05,
63
+ "layer_scale_init_value": 1e-05,
64
+ "max_position_embeddings": 200,
65
+ "model_type": "sam3_lite_text_text_model",
66
+ "num_attention_heads": 4,
67
+ "num_hidden_layers": 2,
68
+ "projection_dim": 32,
69
+ "repmixer_kernel_size": 11,
70
+ "use_repmixer_blocks": false,
71
+ "vocab_size": 49408
72
+ },
73
+ "transformers_version": "5.16.0.dev0",
74
+ "vision_config": {
75
+ "backbone_config": {
76
+ "_name_or_path": "",
77
+ "architectures": null,
78
+ "attention_dropout": 0.0,
79
+ "chunk_size_feed_forward": 0,
80
+ "dtype": null,
81
+ "global_attn_indexes": [
82
+ 0,
83
+ 1
84
+ ],
85
+ "hidden_act": "gelu",
86
+ "hidden_dropout": 0.0,
87
+ "hidden_size": 32,
88
+ "id2label": {
89
+ "0": "LABEL_0",
90
+ "1": "LABEL_1"
91
+ },
92
+ "image_size": 224,
93
+ "initializer_range": 0.02,
94
+ "intermediate_size": 64,
95
+ "is_encoder_decoder": false,
96
+ "label2id": {
97
+ "LABEL_0": 0,
98
+ "LABEL_1": 1
99
+ },
100
+ "layer_norm_eps": 1e-06,
101
+ "layer_scale_init_value": null,
102
+ "model_type": "sam3_vit_model",
103
+ "num_attention_heads": 4,
104
+ "num_channels": 3,
105
+ "num_hidden_layers": 2,
106
+ "output_attentions": false,
107
+ "output_hidden_states": false,
108
+ "patch_size": 14,
109
+ "pretrain_image_size": 336,
110
+ "problem_type": null,
111
+ "return_dict": true,
112
+ "rope_theta": 10000.0,
113
+ "window_size": 8
114
+ },
115
+ "backbone_feature_sizes": [
116
+ [
117
+ 288,
118
+ 288
119
+ ],
120
+ [
121
+ 144,
122
+ 144
123
+ ],
124
+ [
125
+ 72,
126
+ 72
127
+ ]
128
+ ],
129
+ "fpn_hidden_size": 32,
130
+ "hidden_act": "gelu",
131
+ "initializer_range": 0.02,
132
+ "layer_norm_eps": 1e-06,
133
+ "model_type": "sam3_vision_model",
134
+ "scale_factors": [
135
+ 2.0,
136
+ 1.0
137
+ ]
138
+ }
139
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f5edd8afd57cdceb2ca3e4ff9ca278876df676f6057a6864a3c4643524ef768b
3
+ size 7310760
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 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "backend": "tokenizers",
4
+ "bos_token": "<|startoftext|>",
5
+ "do_lower_case": true,
6
+ "eos_token": "<|endoftext|>",
7
+ "errors": "replace",
8
+ "is_local": true,
9
+ "local_files_only": false,
10
+ "max_length": 77,
11
+ "model_max_length": 77,
12
+ "pad_token": "<|endoftext|>",
13
+ "processor_class": "Sam3Processor",
14
+ "tokenizer_class": "CLIPTokenizer",
15
+ "unk_token": "<|endoftext|>"
16
+ }