bert-small-juman-unigram
This model is a fine-tuned version of on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.4490
- Accuracy: 0.6911
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: 0.0001
- train_batch_size: 256
- eval_batch_size: 8
- seed: 42
- distributed_type: multi-GPU
- num_devices: 3
- total_train_batch_size: 768
- total_eval_batch_size: 24
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.01
- num_epochs: 14
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.9849 | 1.0 | 69472 | 1.8385 | 0.6286 |
| 1.8444 | 2.0 | 138944 | 1.6912 | 0.6513 |
| 1.7767 | 3.0 | 208416 | 1.6322 | 0.6610 |
| 1.7357 | 4.0 | 277888 | 1.5931 | 0.6676 |
| 1.709 | 5.0 | 347360 | 1.5636 | 0.6719 |
| 1.6874 | 6.0 | 416832 | 1.5405 | 0.6756 |
| 1.6707 | 7.0 | 486304 | 1.5221 | 0.6786 |
| 1.6511 | 8.0 | 555776 | 1.5061 | 0.6817 |
| 1.636 | 9.0 | 625248 | 1.4933 | 0.6837 |
| 1.6295 | 10.0 | 694720 | 1.4784 | 0.6860 |
| 1.6157 | 11.0 | 764192 | 1.4673 | 0.6879 |
| 1.6027 | 12.0 | 833664 | 1.4605 | 0.6896 |
| 1.5942 | 13.0 | 903136 | 1.4535 | 0.6904 |
| 1.5866 | 14.0 | 972608 | 1.4490 | 0.6911 |
Framework versions
- Transformers 4.19.2
- Pytorch 1.12.0+cu116
- Datasets 2.2.2
- Tokenizers 0.12.1