convnextv2-tiny-22k-384-finetuned-spiderTraining10-1000
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: 0.0815
- Accuracy: 0.9770
- Precision: 0.9760
- Recall: 0.9770
- F1: 0.9763
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 |
|---|---|---|---|---|---|---|---|
| 0.4288 | 1.0 | 125 | 0.2043 | 0.9530 | 0.9533 | 0.9524 | 0.9523 |
| 0.2463 | 2.0 | 250 | 0.1362 | 0.9620 | 0.9619 | 0.9617 | 0.9612 |
| 0.2434 | 3.0 | 375 | 0.0860 | 0.9710 | 0.9699 | 0.9709 | 0.9703 |
| 0.1694 | 4.0 | 500 | 0.0805 | 0.9770 | 0.9763 | 0.9768 | 0.9765 |
| 0.1505 | 5.0 | 625 | 0.0815 | 0.9770 | 0.9760 | 0.9770 | 0.9763 |
Framework versions
- Transformers 4.33.3
- Pytorch 2.0.1+cu117
- Datasets 2.14.5
- Tokenizers 0.13.3