|
--- |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- accuracy |
|
model-index: |
|
- name: danmaku-mlm |
|
results: [] |
|
pipeline_tag: fill-mask |
|
--- |
|
|
|
<!-- 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. --> |
|
|
|
# Danmaku-MLM |
|
|
|
This model is a fine-tuned version of [uer/chinese_roberta_L-8_H-512](https://huggingface.co/uer/chinese_roberta_L-8_H-512) on an [Danmaku Corpus](https://github.com/Chen-X666/bullet-screenCorpus)(2000k raw data) dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 1.1645 |
|
- Accuracy: 0.7780 |
|
|
|
## Model description |
|
|
|
More information needed |
|
|
|
## Intended uses & limitations |
|
|
|
More information needed |
|
|
|
## Training and evaluation data |
|
|
|
Traning Data: https://github.com/Chen-X666/bullet-screenCorpus |
|
|
|
## Training procedure |
|
|
|
### Training hyperparameters |
|
|
|
The following hyperparameters were used during training: |
|
- learning_rate: 5e-05 |
|
- train_batch_size: 64 |
|
- eval_batch_size: 64 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 3.0 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.28.0.dev0 |
|
- Pytorch 1.13.1 |
|
- Datasets 2.6.1 |
|
- Tokenizers 0.11.0 |
