t20spiders-1000
This model is a fine-tuned version of facebook/convnextv2-tiny-22k-384 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.1862
- Accuracy: 0.9400
- Precision: 0.9412
- Recall: 0.9401
- F1: 0.9403
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: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 |
|---|---|---|---|---|---|---|---|
| 0.6303 | 1.0 | 250 | 0.4444 | 0.8729 | 0.8811 | 0.8735 | 0.8734 |
| 0.4303 | 2.0 | 500 | 0.2761 | 0.9135 | 0.9162 | 0.9136 | 0.9136 |
| 0.3779 | 3.0 | 750 | 0.2324 | 0.9245 | 0.9263 | 0.9252 | 0.9248 |
| 0.3067 | 4.0 | 1000 | 0.1997 | 0.9360 | 0.9378 | 0.9358 | 0.9363 |
| 0.2619 | 5.0 | 1250 | 0.1862 | 0.9400 | 0.9412 | 0.9401 | 0.9403 |
Framework versions
- Transformers 4.33.3
- Pytorch 2.0.1+cu117
- Datasets 2.14.5
- Tokenizers 0.13.3