Safetensors
rt_detr_v2
nlivathinos commited on
Commit
72d9495
·
verified ·
1 Parent(s): c2fe333

Upload 3 files

Browse files
Files changed (3) hide show
  1. config.json +155 -0
  2. model.safetensors +3 -0
  3. preprocessor_config.json +26 -0
config.json ADDED
@@ -0,0 +1,155 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation_dropout": 0.0,
3
+ "activation_function": "silu",
4
+ "anchor_image_size": null,
5
+ "architectures": [
6
+ "RTDetrV2ForObjectDetection"
7
+ ],
8
+ "attention_dropout": 0.0,
9
+ "auxiliary_loss": true,
10
+ "backbone": null,
11
+ "backbone_config": {
12
+ "depths": [
13
+ 3,
14
+ 4,
15
+ 6,
16
+ 3
17
+ ],
18
+ "downsample_in_bottleneck": false,
19
+ "downsample_in_first_stage": false,
20
+ "embedding_size": 64,
21
+ "hidden_act": "relu",
22
+ "hidden_sizes": [
23
+ 256,
24
+ 512,
25
+ 1024,
26
+ 2048
27
+ ],
28
+ "layer_type": "bottleneck",
29
+ "model_type": "rt_detr_resnet",
30
+ "num_channels": 3,
31
+ "out_features": [
32
+ "stage2",
33
+ "stage3",
34
+ "stage4"
35
+ ],
36
+ "out_indices": [
37
+ 2,
38
+ 3,
39
+ 4
40
+ ],
41
+ "stage_names": [
42
+ "stem",
43
+ "stage1",
44
+ "stage2",
45
+ "stage3",
46
+ "stage4"
47
+ ]
48
+ },
49
+ "backbone_kwargs": null,
50
+ "batch_norm_eps": 1e-05,
51
+ "box_noise_scale": 1.0,
52
+ "d_model": 256,
53
+ "decoder_activation_function": "relu",
54
+ "decoder_attention_heads": 8,
55
+ "decoder_ffn_dim": 1024,
56
+ "decoder_in_channels": [
57
+ 256,
58
+ 256,
59
+ 256
60
+ ],
61
+ "decoder_layers": 6,
62
+ "decoder_method": "default",
63
+ "decoder_n_levels": 3,
64
+ "decoder_n_points": 4,
65
+ "decoder_offset_scale": 0.5,
66
+ "dropout": 0.0,
67
+ "encode_proj_layers": [
68
+ 2
69
+ ],
70
+ "encoder_activation_function": "gelu",
71
+ "encoder_attention_heads": 8,
72
+ "encoder_ffn_dim": 1024,
73
+ "encoder_hidden_dim": 256,
74
+ "encoder_in_channels": [
75
+ 512,
76
+ 1024,
77
+ 2048
78
+ ],
79
+ "encoder_layers": 1,
80
+ "eos_coefficient": 0.0001,
81
+ "eval_size": null,
82
+ "feat_strides": [
83
+ 8,
84
+ 16,
85
+ 32
86
+ ],
87
+ "focal_loss_alpha": 0.75,
88
+ "focal_loss_gamma": 2.0,
89
+ "freeze_backbone_batch_norms": true,
90
+ "hidden_expansion": 1.0,
91
+ "id2label": {
92
+ "0": "Caption",
93
+ "1": "Footnote",
94
+ "2": "Formula",
95
+ "3": "List-item",
96
+ "4": "Page-footer",
97
+ "5": "Page-header",
98
+ "6": "Picture",
99
+ "7": "Section-header",
100
+ "8": "Table",
101
+ "9": "Text",
102
+ "10": "Title",
103
+ "11": "Document Index",
104
+ "12": "Code",
105
+ "13": "Checkbox-Selected",
106
+ "14": "Checkbox-Unselected",
107
+ "15": "Form",
108
+ "16": "Key-Value Region"
109
+ },
110
+ "initializer_bias_prior_prob": null,
111
+ "initializer_range": 0.01,
112
+ "is_encoder_decoder": true,
113
+ "label2id": {
114
+ "Caption": 0,
115
+ "Checkbox-Selected": 13,
116
+ "Checkbox-Unselected": 14,
117
+ "Code": 12,
118
+ "Document Index": 11,
119
+ "Footnote": 1,
120
+ "Form": 15,
121
+ "Formula": 2,
122
+ "Key-Value Region": 16,
123
+ "List-item": 3,
124
+ "Page-footer": 4,
125
+ "Page-header": 5,
126
+ "Picture": 6,
127
+ "Section-header": 7,
128
+ "Table": 8,
129
+ "Text": 9,
130
+ "Title": 10
131
+ },
132
+ "label_noise_ratio": 0.5,
133
+ "layer_norm_eps": 1e-05,
134
+ "learn_initial_query": false,
135
+ "matcher_alpha": 0.25,
136
+ "matcher_bbox_cost": 5.0,
137
+ "matcher_class_cost": 2.0,
138
+ "matcher_gamma": 2.0,
139
+ "matcher_giou_cost": 2.0,
140
+ "model_type": "rt_detr_v2",
141
+ "normalize_before": false,
142
+ "num_denoising": 100,
143
+ "num_feature_levels": 3,
144
+ "num_queries": 300,
145
+ "positional_encoding_temperature": 10000,
146
+ "torch_dtype": "float32",
147
+ "transformers_version": "4.51.3",
148
+ "use_focal_loss": true,
149
+ "use_pretrained_backbone": false,
150
+ "use_timm_backbone": false,
151
+ "weight_loss_bbox": 5.0,
152
+ "weight_loss_giou": 2.0,
153
+ "weight_loss_vfl": 1.0,
154
+ "with_box_refine": true
155
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b37f85ce9e044df9069471a4b0b764e1be93a44cb7adadfe37e496e928e676f5
3
+ size 171658996
preprocessor_config.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_convert_annotations": true,
3
+ "do_normalize": false,
4
+ "do_pad": false,
5
+ "do_rescale": true,
6
+ "do_resize": true,
7
+ "format": "coco_detection",
8
+ "image_mean": [
9
+ 0.485,
10
+ 0.456,
11
+ 0.406
12
+ ],
13
+ "image_processor_type": "RTDetrImageProcessor",
14
+ "image_std": [
15
+ 0.229,
16
+ 0.224,
17
+ 0.225
18
+ ],
19
+ "pad_size": null,
20
+ "resample": 2,
21
+ "rescale_factor": 0.00392156862745098,
22
+ "size": {
23
+ "height": 640,
24
+ "width": 640
25
+ }
26
+ }