metadata
license: gpl-3.0
base_model: ckiplab/bert-base-chinese
tags:
- generated_from_trainer
model-index:
- name: bert-base-chinese-finetuned-QA-b8
results: []
bert-base-chinese-finetuned-QA-b8
This model is a fine-tuned version of ckiplab/bert-base-chinese on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.3405
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: 3e-05
- train_batch_size: 2
- eval_batch_size: 2
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 8
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 4
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.9325 | 0.14 | 500 | 1.2076 |
| 1.1199 | 0.29 | 1000 | 1.0315 |
| 1.0118 | 0.43 | 1500 | 0.9836 |
| 0.9398 | 0.58 | 2000 | 0.9762 |
| 0.9526 | 0.72 | 2500 | 0.9374 |
| 0.9142 | 0.87 | 3000 | 0.8783 |
| 0.8265 | 1.01 | 3500 | 0.9919 |
| 0.6091 | 1.16 | 4000 | 0.9613 |
| 0.6303 | 1.3 | 4500 | 0.9769 |
| 0.6161 | 1.45 | 5000 | 0.9882 |
| 0.6109 | 1.59 | 5500 | 0.9160 |
| 0.5887 | 1.73 | 6000 | 0.9105 |
| 0.5811 | 1.88 | 6500 | 0.9812 |
| 0.5638 | 2.02 | 7000 | 1.0669 |
| 0.4174 | 2.17 | 7500 | 1.2101 |
| 0.3958 | 2.31 | 8000 | 1.2186 |
| 0.4032 | 2.46 | 8500 | 1.1691 |
| 0.4183 | 2.6 | 9000 | 1.0890 |
| 0.4247 | 2.75 | 9500 | 1.0721 |
| 0.3917 | 2.89 | 10000 | 1.1714 |
| 0.3738 | 3.04 | 10500 | 1.1794 |
| 0.29 | 3.18 | 11000 | 1.2494 |
| 0.326 | 3.32 | 11500 | 1.2822 |
| 0.3076 | 3.47 | 12000 | 1.3214 |
| 0.3071 | 3.61 | 12500 | 1.2968 |
| 0.2797 | 3.76 | 13000 | 1.3410 |
| 0.3192 | 3.9 | 13500 | 1.3405 |
Framework versions
- Transformers 4.34.0
- Pytorch 1.13.1+cu116
- Datasets 2.14.5
- Tokenizers 0.14.1