10-convnextv2-pico-1k-224-finetuned-spiderTraining20-500
This model is a fine-tuned version of facebook/convnextv2-pico-1k-224 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.3910
- Accuracy: 0.8949
- Precision: 0.8942
- Recall: 0.8897
- F1: 0.8904
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 |
|---|---|---|---|---|---|---|---|
| 1.1025 | 1.0 | 80 | 0.9880 | 0.6847 | 0.7201 | 0.6796 | 0.6821 |
| 0.7366 | 2.0 | 160 | 0.6502 | 0.8098 | 0.8198 | 0.8103 | 0.8059 |
| 0.6295 | 3.0 | 240 | 0.5303 | 0.8348 | 0.8410 | 0.8279 | 0.8270 |
| 0.493 | 4.0 | 320 | 0.4666 | 0.8539 | 0.8522 | 0.8533 | 0.8491 |
| 0.3939 | 5.0 | 400 | 0.4831 | 0.8579 | 0.8658 | 0.8503 | 0.8508 |
| 0.3338 | 6.0 | 480 | 0.4551 | 0.8729 | 0.8711 | 0.8670 | 0.8665 |
| 0.2841 | 7.0 | 560 | 0.4357 | 0.8939 | 0.8961 | 0.8931 | 0.8921 |
| 0.2406 | 8.0 | 640 | 0.4074 | 0.8829 | 0.8820 | 0.8760 | 0.8776 |
| 0.2008 | 9.0 | 720 | 0.4074 | 0.8909 | 0.8900 | 0.8872 | 0.8868 |
| 0.2075 | 10.0 | 800 | 0.3910 | 0.8949 | 0.8942 | 0.8897 | 0.8904 |
Framework versions
- Transformers 4.33.3
- Pytorch 2.0.1+cu117
- Datasets 2.14.5
- Tokenizers 0.13.3