detr-r101-cd45rb-4ah-4l-256d
This model is a fine-tuned version of facebook/detr-resnet-101 on the cd45rb dataset. It achieves the following results on the evaluation set:
- Loss: 1.9995
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: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 3.7019 | 1.0 | 4606 | 2.7624 |
| 3.242 | 2.0 | 9212 | 2.4796 |
| 3.0192 | 3.0 | 13818 | 2.2813 |
| 2.863 | 4.0 | 18424 | 2.1890 |
| 2.7702 | 5.0 | 23030 | 2.1246 |
| 2.7018 | 6.0 | 27636 | 2.1121 |
| 2.6499 | 7.0 | 32242 | 2.0616 |
| 2.6082 | 8.0 | 36848 | 2.0321 |
| 2.5774 | 9.0 | 41454 | 2.0095 |
| 2.5589 | 10.0 | 46060 | 1.9995 |
Framework versions
- Transformers 4.28.0
- Pytorch 2.0.1
- Datasets 2.12.0
- Tokenizers 0.13.3