detr-resnet-101_rmsprop_finetuned_food-roboflow
This model is a fine-tuned version of facebook/detr-resnet-101 on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 2.9853
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: 8
- eval_batch_size: 4
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 15
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 5.8836 | 0.77 | 50 | 4.5609 |
| 3.9193 | 1.54 | 100 | 3.5764 |
| 3.2186 | 2.31 | 150 | 3.2685 |
| 2.8952 | 3.08 | 200 | 3.1941 |
| 2.8565 | 3.85 | 250 | 3.1280 |
| 2.7563 | 4.62 | 300 | 3.0448 |
| 2.668 | 5.38 | 350 | 3.0325 |
| 2.6474 | 6.15 | 400 | 2.9764 |
| 2.6463 | 6.92 | 450 | 2.9752 |
| 2.6616 | 7.69 | 500 | 3.0352 |
| 2.5155 | 8.46 | 550 | 2.9928 |
| 2.5778 | 9.23 | 600 | 2.9603 |
| 2.5876 | 10.0 | 650 | 2.9502 |
| 2.513 | 10.77 | 700 | 3.0151 |
| 2.5598 | 11.54 | 750 | 3.0011 |
| 2.491 | 12.31 | 800 | 3.0134 |
| 2.5103 | 13.08 | 850 | 2.9825 |
| 2.5497 | 13.85 | 900 | 3.0169 |
| 2.5738 | 14.62 | 950 | 2.9853 |
Framework versions
- Transformers 4.35.0
- Pytorch 2.1.0+cu118
- Datasets 2.14.6
- Tokenizers 0.14.1