<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
DeTr-TableDetection-3000-images
This model is a fine-tuned version of facebook/detr-resnet-50 on the table_detection_light dataset. It achieves the following results on the evaluation set:
- Loss: 0.3415
- Mean Iou: 0.0220
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 20
Training results
Training Loss | Epoch | Step | Validation Loss | Mean Iou |
---|---|---|---|---|
1.3991 | 1.0 | 188 | 0.8030 | 0.0267 |
0.9905 | 2.0 | 376 | 0.9429 | 0.0262 |
0.6871 | 3.0 | 564 | 1.2024 | 0.0118 |
0.5837 | 4.0 | 752 | 0.5099 | 0.0202 |
0.527 | 5.0 | 940 | 0.4839 | 0.0204 |
0.4847 | 6.0 | 1128 | 0.4542 | 0.0179 |
0.4255 | 7.0 | 1316 | 0.4381 | 0.0227 |
0.3932 | 8.0 | 1504 | 0.4467 | 0.0243 |
0.3846 | 9.0 | 1692 | 0.3817 | 0.0247 |
0.3494 | 10.0 | 1880 | 0.4122 | 0.0261 |
0.3343 | 11.0 | 2068 | 0.3696 | 0.0249 |
0.324 | 12.0 | 2256 | 0.3683 | 0.0242 |
0.3159 | 13.0 | 2444 | 0.3713 | 0.0241 |
0.3079 | 14.0 | 2632 | 0.3563 | 0.0197 |
0.2909 | 15.0 | 2820 | 0.3530 | 0.0237 |
0.279 | 16.0 | 3008 | 0.3513 | 0.0229 |
0.2769 | 17.0 | 3196 | 0.3390 | 0.0211 |
0.2593 | 18.0 | 3384 | 0.3507 | 0.0238 |
0.2533 | 19.0 | 3572 | 0.3370 | 0.0231 |
0.2502 | 20.0 | 3760 | 0.3415 | 0.0220 |
Framework versions
- Transformers 4.26.0
- Pytorch 1.13.1+cu117
- Datasets 2.5.1
- Tokenizers 0.13.2