10-vit-base-patch16-384-finetuned-spiderTraining20-500
This model is a fine-tuned version of google/vit-base-patch16-384 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.3427
- Accuracy: 0.9029
- Precision: 0.9012
- Recall: 0.9032
- F1: 0.9009
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: 0.0005
- train_batch_size: 25
- eval_batch_size: 25
- seed: 42
- distributed_type: multi-GPU
- gradient_accumulation_steps: 4
- total_train_batch_size: 100
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 |
|---|---|---|---|---|---|---|---|
| 0.6925 | 1.0 | 80 | 0.6605 | 0.7938 | 0.8012 | 0.7887 | 0.7869 |
| 0.5869 | 2.0 | 160 | 0.5574 | 0.8298 | 0.8350 | 0.8254 | 0.8214 |
| 0.4858 | 3.0 | 240 | 0.4335 | 0.8689 | 0.8692 | 0.8644 | 0.8644 |
| 0.3921 | 4.0 | 320 | 0.4455 | 0.8739 | 0.8737 | 0.8722 | 0.8699 |
| 0.2915 | 5.0 | 400 | 0.4707 | 0.8629 | 0.8708 | 0.8612 | 0.8571 |
| 0.2727 | 6.0 | 480 | 0.4471 | 0.8819 | 0.8795 | 0.8808 | 0.8777 |
| 0.216 | 7.0 | 560 | 0.3809 | 0.8899 | 0.8879 | 0.8874 | 0.8862 |
| 0.1685 | 8.0 | 640 | 0.3760 | 0.8949 | 0.8938 | 0.8934 | 0.8915 |
| 0.1292 | 9.0 | 720 | 0.3427 | 0.9049 | 0.9034 | 0.9032 | 0.9021 |
| 0.1321 | 10.0 | 800 | 0.3427 | 0.9029 | 0.9012 | 0.9032 | 0.9009 |
Framework versions
- Transformers 4.33.3
- Pytorch 2.0.1+cu117
- Datasets 2.14.5
- Tokenizers 0.13.3