chinese-macbert-base-finetuned-adl_hw1
This model is a fine-tuned version of hfl/chinese-macbert-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1856
- Macro F1: 0.8415
- Micro F1: 0.8403
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: 2e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 100
- num_epochs: 8
Training results
| Training Loss | Epoch | Step | Validation Loss | Macro F1 | Micro F1 |
|---|---|---|---|---|---|
| No log | 1.0 | 169 | 0.3781 | 0.5749 | 0.5684 |
| No log | 2.0 | 338 | 0.2588 | 0.7465 | 0.7483 |
| 0.3392 | 3.0 | 507 | 0.2231 | 0.7889 | 0.7940 |
| 0.3392 | 4.0 | 676 | 0.2012 | 0.8172 | 0.8276 |
| 0.3392 | 5.0 | 845 | 0.1950 | 0.8276 | 0.8255 |
| 0.1585 | 6.0 | 1014 | 0.1874 | 0.8401 | 0.8390 |
| 0.1585 | 7.0 | 1183 | 0.1856 | 0.8415 | 0.8403 |
| 0.1585 | 8.0 | 1352 | 0.1857 | 0.8384 | 0.8367 |
Framework versions
- Transformers 4.50.0
- Pytorch 2.11.0+cu128
- Datasets 2.21.0
- Tokenizers 0.21.4