10-convnextv2-base-22k-384-finetuned-spiderTraining100-1000
This model is a fine-tuned version of facebook/convnextv2-base-22k-384 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.2678
- Accuracy: 0.9313
- Precision: 0.9311
- Recall: 0.9308
- F1: 0.9305
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: 0.0005
- train_batch_size: 27
- eval_batch_size: 27
- seed: 42
- distributed_type: multi-GPU
- gradient_accumulation_steps: 4
- total_train_batch_size: 108
- 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 | Precision | Recall | F1 |
|---|---|---|---|---|---|---|---|
| 1.4003 | 1.0 | 740 | 1.0372 | 0.7086 | 0.7534 | 0.7061 | 0.7057 |
| 1.157 | 2.0 | 1481 | 0.7866 | 0.7733 | 0.8025 | 0.7714 | 0.7726 |
| 0.9605 | 3.0 | 2222 | 0.5709 | 0.8352 | 0.8458 | 0.8346 | 0.8345 |
| 0.7619 | 4.0 | 2963 | 0.4979 | 0.8576 | 0.8659 | 0.8563 | 0.8571 |
| 0.6643 | 5.0 | 3703 | 0.4230 | 0.8807 | 0.8870 | 0.8794 | 0.8797 |
| 0.5426 | 6.0 | 4444 | 0.3959 | 0.8898 | 0.8938 | 0.8889 | 0.8887 |
| 0.5035 | 7.0 | 5185 | 0.3426 | 0.9041 | 0.9063 | 0.9038 | 0.9034 |
| 0.3408 | 8.0 | 5926 | 0.3108 | 0.9188 | 0.9198 | 0.9179 | 0.9180 |
| 0.348 | 9.0 | 6666 | 0.2885 | 0.9238 | 0.9245 | 0.9230 | 0.9230 |
| 0.3402 | 9.99 | 7400 | 0.2678 | 0.9313 | 0.9311 | 0.9308 | 0.9305 |
Framework versions
- Transformers 4.33.3
- Pytorch 2.0.1+cu117
- Datasets 2.14.5
- Tokenizers 0.13.3