10-convnextv2-base-22k-384-finetuned-spiderTraining1000-1000
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: 0.3586
- Accuracy: 0.9180
- Precision: 0.9196
- Recall: 0.9160
- F1: 0.9168
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 |
---|---|---|---|---|---|---|---|
2.0397 | 1.0 | 4064 | 1.4192 | 0.6356 | 0.6956 | 0.6130 | 0.6167 |
1.3997 | 2.0 | 8129 | 0.9638 | 0.7454 | 0.7708 | 0.7320 | 0.7325 |
1.1393 | 3.0 | 12193 | 0.7564 | 0.7973 | 0.8102 | 0.7883 | 0.7884 |
0.9942 | 4.0 | 16258 | 0.6256 | 0.8331 | 0.8464 | 0.8276 | 0.8294 |
0.8572 | 5.0 | 20322 | 0.5610 | 0.8507 | 0.8632 | 0.8441 | 0.8467 |
0.6445 | 6.0 | 24387 | 0.4866 | 0.8730 | 0.8802 | 0.8688 | 0.8697 |
0.5444 | 7.0 | 28451 | 0.4496 | 0.8852 | 0.8909 | 0.8812 | 0.8829 |
0.4955 | 8.0 | 32516 | 0.4241 | 0.8986 | 0.9039 | 0.8952 | 0.8974 |
0.448 | 9.0 | 36580 | 0.3875 | 0.9104 | 0.9133 | 0.9078 | 0.9091 |
0.4109 | 10.0 | 40640 | 0.3586 | 0.9180 | 0.9196 | 0.9160 | 0.9168 |
Framework versions
- Transformers 4.33.3
- Pytorch 2.0.1+cu117
- Datasets 2.14.5
- Tokenizers 0.13.3
- Downloads last month
- 22