detr-r50-mist1-bg-2ah-6l
This model is a fine-tuned version of facebook/detr-resnet-50 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 3.9051
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: 1e-05
- train_batch_size: 4
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 25
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 4.6721 | 1.0 | 115 | 5.0032 |
| 4.4438 | 2.0 | 230 | 4.6797 |
| 4.2953 | 3.0 | 345 | 4.7027 |
| 4.3899 | 4.0 | 460 | 5.4316 |
| 4.3184 | 5.0 | 575 | 4.4125 |
| 4.2749 | 6.0 | 690 | 4.1611 |
| 4.2153 | 7.0 | 805 | 4.6723 |
| 4.0788 | 8.0 | 920 | 4.1266 |
| 4.0752 | 9.0 | 1035 | 4.0529 |
| 4.0073 | 10.0 | 1150 | 4.4483 |
| 4.011 | 11.0 | 1265 | 4.2002 |
| 3.9993 | 12.0 | 1380 | 4.2450 |
| 4.0028 | 13.0 | 1495 | 4.1703 |
| 3.9572 | 14.0 | 1610 | 4.1861 |
| 3.9009 | 15.0 | 1725 | 4.0285 |
| 3.9173 | 16.0 | 1840 | 4.0673 |
| 3.8884 | 17.0 | 1955 | 3.9875 |
| 3.8415 | 18.0 | 2070 | 4.1062 |
| 3.8132 | 19.0 | 2185 | 4.0494 |
| 3.8297 | 20.0 | 2300 | 4.0119 |
| 3.8262 | 21.0 | 2415 | 3.9538 |
| 3.8045 | 22.0 | 2530 | 3.9500 |
| 3.8067 | 23.0 | 2645 | 3.9264 |
| 3.7651 | 24.0 | 2760 | 3.8820 |
| 3.756 | 25.0 | 2875 | 3.9051 |
Framework versions
- Transformers 4.35.0
- Pytorch 2.0.0
- Datasets 2.1.0
- Tokenizers 0.14.1