|
--- |
|
license: apache-2.0 |
|
base_model: hfl/chinese-xlnet-base |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: xlnet-base |
|
results: [] |
|
--- |
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You |
|
should probably proofread and complete it, then remove this comment. --> |
|
|
|
# xlnet-base |
|
|
|
This model is a fine-tuned version of [hfl/chinese-xlnet-base](https://huggingface.co/hfl/chinese-xlnet-base) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 5.7194 |
|
|
|
## 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: 8 |
|
- eval_batch_size: 8 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: cosine |
|
- lr_scheduler_warmup_steps: 500 |
|
- num_epochs: 2 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:----:|:---------------:| |
|
| 8.1116 | 0.11 | 500 | 6.8678 | |
|
| 6.8507 | 0.22 | 1000 | 6.7533 | |
|
| 6.7394 | 0.34 | 1500 | 6.7035 | |
|
| 6.6534 | 0.45 | 2000 | 6.6033 | |
|
| 6.5451 | 0.56 | 2500 | 6.4870 | |
|
| 6.4314 | 0.67 | 3000 | 6.3461 | |
|
| 6.2783 | 0.78 | 3500 | 6.2090 | |
|
| 6.1681 | 0.9 | 4000 | 6.0913 | |
|
| 6.0757 | 1.01 | 4500 | 5.9937 | |
|
| 5.9735 | 1.12 | 5000 | 5.9321 | |
|
| 5.9025 | 1.23 | 5500 | 5.8552 | |
|
| 5.8424 | 1.34 | 6000 | 5.8166 | |
|
| 5.804 | 1.45 | 6500 | 5.7849 | |
|
| 5.7535 | 1.57 | 7000 | 5.7420 | |
|
| 5.7674 | 1.68 | 7500 | 5.7311 | |
|
| 5.7613 | 1.79 | 8000 | 5.7269 | |
|
| 5.7322 | 1.9 | 8500 | 5.7194 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.36.2 |
|
- Pytorch 2.1.0+cu121 |
|
- Datasets 2.16.0 |
|
- Tokenizers 0.15.0 |
|
|
