vit-base-classification-new
This model is a fine-tuned version of google/vit-base-patch16-224-in21k on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0101
- F1: 0.9891
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: 0.0002
- train_batch_size: 32
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 6
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | F1 |
|---|---|---|---|---|
| 0.4961 | 1.0 | 212 | 0.4091 | 0.8536 |
| 0.1776 | 2.0 | 424 | 0.1429 | 0.9425 |
| 0.0619 | 3.0 | 636 | 0.0713 | 0.9783 |
| 0.0281 | 4.0 | 848 | 0.0239 | 0.9942 |
| 0.0125 | 5.0 | 1060 | 0.0138 | 0.9931 |
| 0.0067 | 6.0 | 1272 | 0.0101 | 0.9891 |
Framework versions
- Transformers 4.40.2
- Pytorch 2.2.1+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1