convnext-base-224-augmented-sdc
This model is a fine-tuned version of facebook/convnext-base-224 on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 0.5343
- Accuracy: 0.8634
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: 128
- eval_batch_size: 128
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 512
- 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 |
---|---|---|---|---|
2.9671 | 1.0 | 69 | 2.6640 | 0.2875 |
1.8274 | 2.0 | 138 | 1.6218 | 0.6078 |
1.3433 | 3.0 | 207 | 1.1754 | 0.7102 |
1.0912 | 4.0 | 276 | 0.9481 | 0.7572 |
0.9112 | 5.0 | 345 | 0.7916 | 0.7980 |
0.8505 | 6.0 | 414 | 0.6876 | 0.8299 |
0.7691 | 7.0 | 483 | 0.6165 | 0.8483 |
0.7163 | 8.0 | 552 | 0.5735 | 0.8557 |
0.6789 | 9.0 | 621 | 0.5471 | 0.8608 |
0.6531 | 10.0 | 690 | 0.5343 | 0.8634 |
Framework versions
- Transformers 4.39.1
- Pytorch 2.2.2+cu118
- Datasets 2.18.0
- Tokenizers 0.15.2
- Downloads last month
- 4