resnet-50-finetuned-FER2013CKPlus-0.003
This model is a fine-tuned version of Celal11/resnet-50-finetuned-FER2013-0.003 on the image_folder dataset. It achieves the following results on the evaluation set:
- Loss: 0.0073
- Accuracy: 1.0
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: 0.003
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.8084 | 0.97 | 27 | 0.2004 | 0.9289 |
| 0.362 | 1.97 | 54 | 0.0828 | 0.9848 |
| 0.2972 | 2.97 | 81 | 0.0185 | 0.9949 |
| 0.1917 | 3.97 | 108 | 0.0132 | 1.0 |
| 0.1572 | 4.97 | 135 | 0.0073 | 1.0 |
Framework versions
- Transformers 4.20.1
- Pytorch 1.11.0
- Datasets 2.1.0
- Tokenizers 0.12.1