Intro
Just a ML-2 HSE course homework done by Zaryvnykh Amaliya, DSBA201
Results
This model is a fine-tuned version of google/vit-base-patch16-224-in21k on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.0381
- Accuracy: 0.9867
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: 16
- eval_batch_size: 16
- seed: 1337
- gradient_accumulation_steps: 16
- total_train_batch_size: 256
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.0984 | 0.98 | 26 | 0.0847 | 0.9725 |
| 0.0493 | 2.0 | 53 | 0.0480 | 0.9842 |
| 0.0407 | 2.97 | 79 | 0.0456 | 0.9867 |
| 0.033 | 3.99 | 106 | 0.0400 | 0.9858 |
| 0.0261 | 4.89 | 130 | 0.0388 | 0.9892 |
Framework versions
- Transformers 4.35.0
- Pytorch 2.1.0+cu121
- Datasets 2.14.6
- Tokenizers 0.14.1