10-convnextv2-nano-22k-384-finetuned-spiderTraining20-500
This model is a fine-tuned version of facebook/convnextv2-nano-22k-384 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.2966
- Accuracy: 0.9109
- Precision: 0.9058
- Recall: 0.9065
- F1: 0.9057
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: 25
- eval_batch_size: 25
- seed: 42
- distributed_type: multi-GPU
- gradient_accumulation_steps: 4
- total_train_batch_size: 100
- 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 |
|---|---|---|---|---|---|---|---|
| 1.7156 | 1.0 | 80 | 1.4314 | 0.6316 | 0.6182 | 0.6246 | 0.6155 |
| 0.7565 | 2.0 | 160 | 0.6340 | 0.8168 | 0.8213 | 0.8074 | 0.8095 |
| 0.5802 | 3.0 | 240 | 0.4632 | 0.8589 | 0.8566 | 0.8545 | 0.8539 |
| 0.4767 | 4.0 | 320 | 0.4006 | 0.8759 | 0.8748 | 0.8710 | 0.8708 |
| 0.3648 | 5.0 | 400 | 0.3529 | 0.8999 | 0.8976 | 0.8965 | 0.8960 |
| 0.3623 | 6.0 | 480 | 0.3326 | 0.9059 | 0.9030 | 0.9031 | 0.9024 |
| 0.3238 | 7.0 | 560 | 0.3178 | 0.8939 | 0.8910 | 0.8889 | 0.8892 |
| 0.2975 | 8.0 | 640 | 0.3016 | 0.9079 | 0.9037 | 0.9029 | 0.9028 |
| 0.2852 | 9.0 | 720 | 0.3090 | 0.9029 | 0.8979 | 0.8991 | 0.8974 |
| 0.2893 | 10.0 | 800 | 0.2966 | 0.9109 | 0.9058 | 0.9065 | 0.9057 |
Framework versions
- Transformers 4.33.3
- Pytorch 2.0.1+cu117
- Datasets 2.14.5
- Tokenizers 0.13.3