metadata
license: apache-2.0
base_model: microsoft/swinv2-base-patch4-window12to16-192to256-22kto1k-ft
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: No-Augmentation-swinv2-base
results: []
No-Augmentation-swinv2-base
This model is a fine-tuned version of microsoft/swinv2-base-patch4-window12to16-192to256-22kto1k-ft on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.9741
- Accuracy: 0.7589
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: 5e-05
- 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: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 2.3968 | 0.96 | 18 | 1.3103 | 0.5731 |
| 0.8906 | 1.97 | 37 | 0.9292 | 0.6838 |
| 0.4531 | 2.99 | 56 | 0.8277 | 0.7352 |
| 0.2294 | 4.0 | 75 | 0.8021 | 0.7273 |
| 0.1208 | 4.96 | 93 | 0.7764 | 0.7589 |
| 0.0745 | 5.97 | 112 | 0.8899 | 0.7549 |
| 0.0733 | 6.99 | 131 | 0.9730 | 0.7549 |
| 0.0381 | 8.0 | 150 | 0.9304 | 0.7747 |
| 0.0151 | 8.96 | 168 | 0.9733 | 0.7510 |
| 0.0233 | 9.6 | 180 | 0.9741 | 0.7589 |
Framework versions
- Transformers 4.39.3
- Pytorch 2.1.2
- Datasets 2.19.1
- Tokenizers 0.15.2