Hugging Face's logo

nasiraziz321
/
rtdetr-r50-bench-finetune

Object Detection
Transformers
TensorBoard
Safetensors
rt_detr
Generated from Trainer
Model card Files
xet
Metrics Community
nasiraziz321 commited on
Commit
8d0bd24
·
verified ·
1 Parent(s): adb9e2a

End of training

Browse files
Files changed (7) hide show
  1. README.md +94 -0
  2. config.json +133 -0
  3. model.safetensors +3 -0
  4. preprocessor_config.json +26 -0
  5. runs/Jul07_05-55-06_0f12436c74cb/events.out.tfevents.1751867711.0f12436c74cb.614.0 +3 -0
  6. runs/Jul07_05-55-06_0f12436c74cb/events.out.tfevents.1751868925.0f12436c74cb.614.1 +3 -0
  7. training_args.bin +3 -0
README.md ADDED
@@ -0,0 +1,94 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: transformers
3
+ license: apache-2.0
4
+ base_model: PekingU/rtdetr_r50vd_coco_o365
5
+ tags:
6
+ - generated_from_trainer
7
+ model-index:
8
+ - name: rtdetr-r50-bench-finetune
9
+ results: []
10
+ ---
11
+
12
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
13
+ should probably proofread and complete it, then remove this comment. -->
14
+
15
+ # rtdetr-r50-bench-finetune
16
+
17
+ This model is a fine-tuned version of [PekingU/rtdetr_r50vd_coco_o365](https://huggingface.co/PekingU/rtdetr_r50vd_coco_o365) on an unknown dataset.
18
+ It achieves the following results on the evaluation set:
19
+ - Loss: 15.1486
20
+ - Map: 0.217
21
+ - Map 50: 0.3852
22
+ - Map 75: 0.1895
23
+ - Map Small: 0.206
24
+ - Map Medium: 0.2572
25
+ - Map Large: 0.0515
26
+ - Mar 1: 0.2172
27
+ - Mar 10: 0.3591
28
+ - Mar 100: 0.3796
29
+ - Mar Small: 0.3184
30
+ - Mar Medium: 0.4616
31
+ - Mar Large: 0.05
32
+ - Map Elbows-shoulders-bar-athlete: -1.0
33
+ - Mar 100 Elbows-shoulders-bar-athlete: -1.0
34
+ - Map Barbell: 0.1248
35
+ - Mar 100 Barbell: 0.305
36
+ - Map Benchpresschest: -1.0
37
+ - Mar 100 Benchpresschest: -1.0
38
+ - Map Benchpressshoulder: 0.0
39
+ - Mar 100 Benchpressshoulder: 0.0
40
+ - Map Elbow: 0.01
41
+ - Mar 100 Elbow: 0.3238
42
+ - Map Hand: 0.3141
43
+ - Mar 100 Hand: 0.5248
44
+ - Map Head: 0.636
45
+ - Mar 100 Head: 0.7447
46
+
47
+ ## Model description
48
+
49
+ More information needed
50
+
51
+ ## Intended uses & limitations
52
+
53
+ More information needed
54
+
55
+ ## Training and evaluation data
56
+
57
+ More information needed
58
+
59
+ ## Training procedure
60
+
61
+ ### Training hyperparameters
62
+
63
+ The following hyperparameters were used during training:
64
+ - learning_rate: 5e-05
65
+ - train_batch_size: 8
66
+ - eval_batch_size: 8
67
+ - seed: 42
68
+ - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
69
+ - lr_scheduler_type: linear
70
+ - lr_scheduler_warmup_steps: 300
71
+ - num_epochs: 10
72
+
73
+ ### Training results
74
+
75
+ | Training Loss | Epoch | Step | Validation Loss | Map | Map 50 | Map 75 | Map Small | Map Medium | Map Large | Mar 1 | Mar 10 | Mar 100 | Mar Small | Mar Medium | Mar Large | Map Elbows-shoulders-bar-athlete | Mar 100 Elbows-shoulders-bar-athlete | Map Barbell | Mar 100 Barbell | Map Benchpresschest | Mar 100 Benchpresschest | Map Benchpressshoulder | Mar 100 Benchpressshoulder | Map Elbow | Mar 100 Elbow | Map Hand | Mar 100 Hand | Map Head | Mar 100 Head |
76
+ |:-------------:|:-----:|:----:|:---------------:|:------:|:------:|:------:|:---------:|:----------:|:---------:|:------:|:------:|:-------:|:---------:|:----------:|:---------:|:--------------------------------:|:------------------------------------:|:-----------:|:---------------:|:-------------------:|:-----------------------:|:----------------------:|:--------------------------:|:---------:|:-------------:|:--------:|:------------:|:--------:|:------------:|
77
+ | No log | 1.0 | 37 | 58.1399 | 0.0001 | 0.0001 | 0.0001 | 0.0 | 0.0002 | 0.0 | 0.0021 | 0.0028 | 0.0044 | 0.002 | 0.0059 | 0.0 | -1.0 | -1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.009 | 0.0 | 0.005 | 0.0003 | 0.0123 |
78
+ | No log | 2.0 | 74 | 42.4180 | 0.0038 | 0.0056 | 0.0054 | 0.0001 | 0.0103 | 0.0 | 0.0062 | 0.0156 | 0.0156 | 0.0021 | 0.0218 | 0.0 | -1.0 | -1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0007 | 0.0256 | 0.022 | 0.0679 |
79
+ | No log | 3.0 | 111 | 34.4996 | 0.015 | 0.0217 | 0.0178 | 0.0002 | 0.0327 | 0.0 | 0.0245 | 0.0383 | 0.0385 | 0.0032 | 0.0552 | 0.0 | -1.0 | -1.0 | 0.0003 | 0.0043 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0031 | 0.0525 | 0.0867 | 0.1741 |
80
+ | No log | 4.0 | 148 | 29.1792 | 0.0254 | 0.0435 | 0.0239 | 0.0083 | 0.0486 | 0.0014 | 0.0349 | 0.0626 | 0.0626 | 0.0245 | 0.0885 | 0.02 | -1.0 | -1.0 | 0.0004 | 0.0021 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0587 | 0.2044 | 0.0933 | 0.1691 |
81
+ | No log | 5.0 | 185 | 22.1037 | 0.0867 | 0.1572 | 0.0786 | 0.06 | 0.1234 | 0.0 | 0.0902 | 0.1323 | 0.1324 | 0.1011 | 0.1681 | 0.0 | -1.0 | -1.0 | 0.0022 | 0.0209 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0108 | 0.2576 | 0.4181 | 0.2605 | 0.3444 |
82
+ | No log | 6.0 | 222 | 18.5807 | 0.1367 | 0.2393 | 0.13 | 0.1254 | 0.1771 | 0.0027 | 0.1285 | 0.2144 | 0.2357 | 0.2282 | 0.3049 | 0.04 | -1.0 | -1.0 | 0.0143 | 0.1235 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0005 | 0.1175 | 0.3634 | 0.5731 | 0.4419 | 0.6 |
83
+ | No log | 7.0 | 259 | 17.0227 | 0.1642 | 0.2798 | 0.1596 | 0.1853 | 0.2005 | 0.0153 | 0.1531 | 0.2538 | 0.2663 | 0.2964 | 0.3338 | 0.04 | -1.0 | -1.0 | 0.0456 | 0.1674 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0014 | 0.1687 | 0.3894 | 0.5825 | 0.5486 | 0.679 |
84
+ | No log | 8.0 | 296 | 15.6265 | 0.1849 | 0.3282 | 0.1716 | 0.2236 | 0.2215 | 0.0208 | 0.1801 | 0.2874 | 0.2967 | 0.3501 | 0.3619 | 0.02 | -1.0 | -1.0 | 0.1076 | 0.2984 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0017 | 0.197 | 0.4029 | 0.5838 | 0.5973 | 0.7012 |
85
+ | No log | 9.0 | 333 | 15.1906 | 0.1871 | 0.3393 | 0.166 | 0.2442 | 0.2201 | 0.0401 | 0.1826 | 0.3077 | 0.3425 | 0.4268 | 0.3931 | 0.1 | -1.0 | -1.0 | 0.1194 | 0.3396 | 0.0 | 0.0 | 0.002 | 0.05 | 0.0071 | 0.3747 | 0.3975 | 0.5638 | 0.5969 | 0.7272 |
86
+ | No log | 10.0 | 370 | 14.8474 | 0.1954 | 0.3502 | 0.1848 | 0.2464 | 0.2337 | 0.0208 | 0.1818 | 0.2982 | 0.311 | 0.3738 | 0.3728 | 0.02 | -1.0 | -1.0 | 0.1522 | 0.331 | 0.0 | 0.0 | 0.0 | 0.0 | 0.006 | 0.2627 | 0.4073 | 0.565 | 0.6066 | 0.7074 |
87
+
88
+
89
+ ### Framework versions
90
+
91
+ - Transformers 4.54.0.dev0
92
+ - Pytorch 2.6.0+cu124
93
+ - Datasets 3.6.0
94
+ - Tokenizers 0.21.2
config.json ADDED
@@ -0,0 +1,133 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation_dropout": 0.0,
3
+ "activation_function": "silu",
4
+ "anchor_image_size": null,
5
+ "architectures": [
6
+ "RTDetrForObjectDetection"
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_n_points": 4,
63
+ "disable_custom_kernels": true,
64
+ "dropout": 0.0,
65
+ "encode_proj_layers": [
66
+ 2
67
+ ],
68
+ "encoder_activation_function": "gelu",
69
+ "encoder_attention_heads": 8,
70
+ "encoder_ffn_dim": 1024,
71
+ "encoder_hidden_dim": 256,
72
+ "encoder_in_channels": [
73
+ 512,
74
+ 1024,
75
+ 2048
76
+ ],
77
+ "encoder_layers": 1,
78
+ "eos_coefficient": 0.0001,
79
+ "eval_size": null,
80
+ "feat_strides": [
81
+ 8,
82
+ 16,
83
+ 32
84
+ ],
85
+ "focal_loss_alpha": 0.75,
86
+ "focal_loss_gamma": 2.0,
87
+ "freeze_backbone_batch_norms": true,
88
+ "hidden_expansion": 1.0,
89
+ "id2label": {
90
+ "0": "elbows-shoulders-bar-athlete",
91
+ "1": "barbell",
92
+ "2": "benchpresschest",
93
+ "3": "benchpressshoulder",
94
+ "4": "elbow",
95
+ "5": "hand",
96
+ "6": "head"
97
+ },
98
+ "initializer_bias_prior_prob": null,
99
+ "initializer_range": 0.01,
100
+ "is_encoder_decoder": true,
101
+ "label2id": {
102
+ "barbell": 1,
103
+ "benchpresschest": 2,
104
+ "benchpressshoulder": 3,
105
+ "elbow": 4,
106
+ "elbows-shoulders-bar-athlete": 0,
107
+ "hand": 5,
108
+ "head": 6
109
+ },
110
+ "label_noise_ratio": 0.5,
111
+ "layer_norm_eps": 1e-05,
112
+ "learn_initial_query": false,
113
+ "matcher_alpha": 0.25,
114
+ "matcher_bbox_cost": 5.0,
115
+ "matcher_class_cost": 2.0,
116
+ "matcher_gamma": 2.0,
117
+ "matcher_giou_cost": 2.0,
118
+ "model_type": "rt_detr",
119
+ "normalize_before": false,
120
+ "num_denoising": 100,
121
+ "num_feature_levels": 3,
122
+ "num_queries": 300,
123
+ "positional_encoding_temperature": 10000,
124
+ "torch_dtype": "float32",
125
+ "transformers_version": "4.54.0.dev0",
126
+ "use_focal_loss": true,
127
+ "use_pretrained_backbone": false,
128
+ "use_timm_backbone": false,
129
+ "weight_loss_bbox": 5.0,
130
+ "weight_loss_giou": 2.0,
131
+ "weight_loss_vfl": 1.0,
132
+ "with_box_refine": true
133
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:403c6fe2b87c90ee9043ff41c4f75fc2e28b8a4eae6cc29941aea9155412cf61
3
+ size 171575780
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": 480,
24
+ "width": 480
25
+ }
26
+ }
runs/Jul07_05-55-06_0f12436c74cb/events.out.tfevents.1751867711.0f12436c74cb.614.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ce8208b043e0db8889eb7d3441ba47b93312dae9ad6d8995db324c4728da3d1e
3
+ size 24928
runs/Jul07_05-55-06_0f12436c74cb/events.out.tfevents.1751868925.0f12436c74cb.614.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:909a3f0581f6af0f5a701cb441b427606265b3b26c24a504361294ad997b7829
3
+ size 1838
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:23517852f381c898379bf70b954516c0668654a8721de3eabf587dd611735e2c
3
+ size 5368