File size: 8,355 Bytes
f395faf |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 |
---
language: Chinese
datasets: CLUECorpusSmall
widget:
- text: "北京是[MASK]国的首都。"
---
# Chinese RoBERTa Miniatures
## Model description
This is the set of 24 Chinese RoBERTa models pre-trained by [UER-py](https://www.aclweb.org/anthology/D19-3041.pdf).
[Turc et al.](https://arxiv.org/abs/1908.08962) have shown that the standard BERT recipe is effective on a wide range of model sizes. Following their paper, we released the 24 Chinese RoBERTa models. In order to facilitate users to reproduce the results, we used the publicly available corpus and provided all training details.
You can download the 24 Chinese RoBERTa miniatures either from the [UER-py Github page](https://github.com/dbiir/UER-py/), or via HuggingFace from the links below:
| | H=128 | H=256 | H=512 | H=768 |
| -------- | :-----------------------: | :-----------------------: | :-------------------------: | :-------------------------: |
| **L=2** | [**2/128 (Tiny)**][2_128] | [2/256] | [2/512] | [2/768] |
| **L=4** | [4/128] | [**4/256 (Mini)**][4_256] | [**4/512 (Small)**][4_512] | [4/768] |
| **L=6** | [6/128] | [6/256] | [6/512] | [6/768] |
| **L=8** | [8/128] | [8/256] | [**8/512 (Medium)**][8_512] | [8/768] |
| **L=10** | [10/128] | [10/256] | [10/512] | [10/768] |
| **L=12** | [12/128] | [12/256] | [12/512] | [**12/768 (Base)**][12_768] |
Here are scores on the devlopment set of six Chinese tasks:
| Model | Score | douban | chnsenticorp | lcqmc | tnews(CLUE) | iflytek(CLUE) | ocnli(CLUE) |
| -------------- | :---: | :----: | :----------: | :---: | :---------: | :-----------: | :---------: |
| RoBERTa-Tiny | 72.3 | 83.0 | 91.4 | 81.8 | 62.0 | 55.0 | 60.3 |
| RoBERTa-Mini | 75.7 | 84.8 | 93.7 | 86.1 | 63.9 | 58.3 | 67.4 |
| RoBERTa-Small | 76.8 | 86.5 | 93.4 | 86.5 | 65.1 | 59.4 | 69.7 |
| RoBERTa-Medium | 77.8 | 87.6 | 94.8 | 88.1 | 65.6 | 59.5 | 71.2 |
| RoBERTa-Base | 79.5 | 89.1 | 95.2 | 89.2 | 67.0 | 60.9 | 75.5 |
For each task, we selected the best fine-tuning hyperparameters from the lists below, and trained with the sequence length of 128:
- epochs: 3, 5, 8
- batch sizes: 32, 64
- learning rates: 3e-5, 1e-4, 3e-4
## How to use
You can use this model directly with a pipeline for masked language modeling (take the case of RoBERTa-Medium):
```python
>>> from transformers import pipeline
>>> unmasker = pipeline('fill-mask', model='uer/chinese_roberta_L-8_H-512')
>>> unmasker("中国的首都是[MASK]京。")
[
{'sequence': '[CLS] 中 国 的 首 都 是 北 京 。 [SEP]',
'score': 0.8701988458633423,
'token': 1266,
'token_str': '北'},
{'sequence': '[CLS] 中 国 的 首 都 是 南 京 。 [SEP]',
'score': 0.1194809079170227,
'token': 1298,
'token_str': '南'},
{'sequence': '[CLS] 中 国 的 首 都 是 东 京 。 [SEP]',
'score': 0.0037803512532263994,
'token': 691,
'token_str': '东'},
{'sequence': '[CLS] 中 国 的 首 都 是 普 京 。 [SEP]',
'score': 0.0017127094324678183,
'token': 3249,
'token_str': '普'},
{'sequence': '[CLS] 中 国 的 首 都 是 望 京 。 [SEP]',
'score': 0.001687526935711503,
'token': 3307,
'token_str': '望'}
]
```
Here is how to use this model to get the features of a given text in PyTorch:
```python
from transformers import BertTokenizer, BertModel
tokenizer = BertTokenizer.from_pretrained('uer/chinese_roberta_L-8_H-512')
model = BertModel.from_pretrained("uer/chinese_roberta_L-8_H-512")
text = "用你喜欢的任何文本替换我。"
encoded_input = tokenizer(text, return_tensors='pt')
output = model(**encoded_input)
```
and in TensorFlow:
```python
from transformers import BertTokenizer, TFBertModel
tokenizer = BertTokenizer.from_pretrained('uer/chinese_roberta_L-8_H-512')
model = TFBertModel.from_pretrained("uer/chinese_roberta_L-8_H-512")
text = "用你喜欢的任何文本替换我。"
encoded_input = tokenizer(text, return_tensors='tf')
output = model(encoded_input)
```
## Training data
[CLUECorpusSmall](https://github.com/CLUEbenchmark/CLUECorpus2020/) is used as training data. We found that models pre-trained on CLUECorpusSmall outperform those pre-trained on CLUECorpus2020, although CLUECorpus2020 is much larger than CLUECorpusSmall.
## Training procedure
Models are pre-trained by [UER-py](https://github.com/dbiir/UER-py/) on [Tencent Cloud TI-ONE](https://cloud.tencent.com/product/tione/). We pre-train 1,000,000 steps with a sequence length of 128 and then pre-train 250,000 additional steps with a sequence length of 512.
Taking the case of RoBERTa-Medium
Stage1:
```
python3 preprocess.py --corpus_path corpora/cluecorpussmall.txt \
--vocab_path models/google_zh_vocab.txt \
--dataset_path cluecorpussmall_seq128_dataset.pt \
--processes_num 32 --seq_length 128 \
--dynamic_masking --target mlm
```
```
python3 pretrain.py --dataset_path cluecorpussmall_seq128_dataset.pt \
--vocab_path models/google_zh_vocab.txt \
--config_path models/bert_medium_config.json \
--output_model_path models/cluecorpussmall_roberta_medium_seq128_model.bin \
--world_size 8 --gpu_ranks 0 1 2 3 4 5 6 7 \
--total_steps 1000000 --save_checkpoint_steps 100000 --report_steps 50000 \
--learning_rate 1e-4 --batch_size 64 \
--tie_weights --embedding word_pos_seg --encoder transformer --mask fully_visible --target mlm
```
Stage2:
```
python3 preprocess.py --corpus_path corpora/cluecorpussmall.txt \
--vocab_path models/google_zh_vocab.txt \
--dataset_path cluecorpussmall_seq512_dataset.pt \
--processes_num 32 --seq_length 512 \
--dynamic_masking --target mlm
```
```
python3 pretrain.py --dataset_path cluecorpussmall_seq512_dataset.pt \
--pretrained_model_path models/cluecorpussmall_roberta_medium_seq128_model.bin-1000000 \
--vocab_path models/google_zh_vocab.txt \
--config_path models/bert_medium_config.json \
--output_model_path models/cluecorpussmall_roberta_medium_seq512_model.bin \
--world_size 8 --gpu_ranks 0 1 2 3 4 5 6 7 \
--total_steps 250000 --save_checkpoint_steps 50000 --report_steps 10000 \
--learning_rate 5e-5 --batch_size 16 \
--tie_weights --embedding word_pos_seg --encoder transformer --mask fully_visible --target mlm
```
Finally, we convert the pre-trained model into Huggingface's format:
```
python3 scripts/convert_bert_from_uer_to_huggingface.py --input_model_path models/cluecorpussmall_roberta_medium_seq512_model.bin-250000 \
--output_model_path pytorch_model.bin \
--layers_num 8 --target mlm
```
### BibTeX entry and citation info
```
@article{zhao2019uer,
title={UER: An Open-Source Toolkit for Pre-training Models},
author={Zhao, Zhe and Chen, Hui and Zhang, Jinbin and Zhao, Xin and Liu, Tao and Lu, Wei and Chen, Xi and Deng, Haotang and Ju, Qi and Du, Xiaoyong},
journal={EMNLP-IJCNLP 2019},
pages={241},
year={2019}
}
```
[2_128]: https://huggingface.co/uer/chinese_roberta_L-2_H-128
[4_256]: https://huggingface.co/uer/chinese_roberta_L-4_H-256
[4_512]: https://huggingface.co/uer/chinese_roberta_L-4_H-512
[8_512]: https://huggingface.co/uer/chinese_roberta_L-8_H-512
[12_768]: https://huggingface.co/uer/chinese_roberta_L-12_H-768 |