vit-base-patch16-224-RU5-10-8
This model is a fine-tuned version of google/vit-base-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 0.6773
- Accuracy: 0.7833
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: 5.5e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.05
- num_epochs: 10
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy |
---|---|---|---|---|
1.3605 | 0.95 | 14 | 1.2370 | 0.5167 |
1.2314 | 1.97 | 29 | 1.0511 | 0.6833 |
0.968 | 2.98 | 44 | 0.8919 | 0.65 |
0.8135 | 4.0 | 59 | 0.7702 | 0.7667 |
0.616 | 4.95 | 73 | 0.7533 | 0.75 |
0.5167 | 5.97 | 88 | 0.6773 | 0.7833 |
0.4063 | 6.98 | 103 | 0.6974 | 0.75 |
0.3401 | 8.0 | 118 | 0.7438 | 0.75 |
0.3007 | 8.95 | 132 | 0.6646 | 0.7833 |
0.3154 | 9.49 | 140 | 0.6819 | 0.7833 |
Framework versions
- Transformers 4.36.2
- Pytorch 2.1.2+cu118
- Datasets 2.16.1
- Tokenizers 0.15.0
- Downloads last month
- 0
Finetuned from
Evaluation results
- Accuracy on imagefoldervalidation set self-reported0.783