<!-- 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-1000-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: 0.5143
- Mean Iou: 0.0242
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 |
---|---|---|---|---|
No log | 1.0 | 63 | 0.8696 | 0.0215 |
0.9227 | 2.0 | 126 | 0.7547 | 0.0245 |
0.9227 | 3.0 | 189 | 0.7170 | 0.0211 |
0.6775 | 4.0 | 252 | 0.8319 | 0.0222 |
0.6801 | 5.0 | 315 | 0.6943 | 0.0212 |
0.6801 | 6.0 | 378 | 0.6622 | 0.0252 |
0.604 | 7.0 | 441 | 0.6043 | 0.0234 |
0.5467 | 8.0 | 504 | 0.7404 | 0.0249 |
0.5467 | 9.0 | 567 | 0.6755 | 0.0242 |
0.4347 | 10.0 | 630 | 0.5507 | 0.0232 |
0.4347 | 11.0 | 693 | 0.6633 | 0.0277 |
0.4202 | 12.0 | 756 | 0.5941 | 0.0256 |
0.3508 | 13.0 | 819 | 0.5387 | 0.0238 |
0.3508 | 14.0 | 882 | 0.5381 | 0.0256 |
0.3223 | 15.0 | 945 | 0.5646 | 0.0254 |
0.3058 | 16.0 | 1008 | 0.5460 | 0.0213 |
0.3058 | 17.0 | 1071 | 0.5589 | 0.0264 |
0.2861 | 18.0 | 1134 | 0.5423 | 0.0257 |
0.2861 | 19.0 | 1197 | 0.5207 | 0.0248 |
0.2705 | 20.0 | 1260 | 0.5143 | 0.0242 |
Framework versions
- Transformers 4.26.0
- Pytorch 1.13.1+cu117
- Datasets 2.5.1
- Tokenizers 0.13.2