Update README.md
Browse files
README.md
CHANGED
|
@@ -36,25 +36,11 @@ The model is pre-trained by [UER-py](https://github.com/dbiir/UER-py/) on [Tence
|
|
| 36 |
|
| 37 |
|
| 38 |
```
|
| 39 |
-
python3 preprocess.py --corpus_path corpora/cluecorpussmall.txt
|
| 40 |
-
--vocab_path models/google_zh_vocab.txt \
|
| 41 |
-
--dataset_path cluecorpussmall_bart_seq512_dataset.pt \
|
| 42 |
-
--processes_num 32 --seq_length 512 \
|
| 43 |
-
--dynamic_masking --target bart
|
| 44 |
```
|
| 45 |
|
| 46 |
```
|
| 47 |
-
python3 pretrain.py --dataset_path cluecorpussmall_bart_seq512_dataset.pt
|
| 48 |
-
--vocab_path models/google_zh_vocab.txt \
|
| 49 |
-
--config_path models/bart/base_config.json \
|
| 50 |
-
--output_model_path models/cluecorpussmall_bart_base_seq512_model.bin \
|
| 51 |
-
--world_size 8 --gpu_ranks 0 1 2 3 4 5 6 7 \
|
| 52 |
-
--total_steps 1000000 --save_checkpoint_steps 100000 --report_steps 50000 \
|
| 53 |
-
--learning_rate 1e-4 --batch_size 16 \
|
| 54 |
-
--span_masking --span_max_length 3 \
|
| 55 |
-
--embedding word_pos --tgt_embedding word_pos \
|
| 56 |
-
--encoder transformer --mask fully_visible --decoder transformer \
|
| 57 |
-
--target bart --tie_weights --has_lmtarget_bias
|
| 58 |
```
|
| 59 |
|
| 60 |
Finally, we convert the pre-trained model into Huggingface's format:
|
|
|
|
| 36 |
|
| 37 |
|
| 38 |
```
|
| 39 |
+
python3 preprocess.py --corpus_path corpora/cluecorpussmall.txt \\n --vocab_path models/google_zh_vocab.txt \\n --dataset_path cluecorpussmall_bart_seq512_dataset.pt \\n --processes_num 32 --seq_length 512 \\n --target bart
|
|
|
|
|
|
|
|
|
|
|
|
|
| 40 |
```
|
| 41 |
|
| 42 |
```
|
| 43 |
+
python3 pretrain.py --dataset_path cluecorpussmall_bart_seq512_dataset.pt \\n --vocab_path models/google_zh_vocab.txt \\n --config_path models/bart/base_config.json \\n --output_model_path models/cluecorpussmall_bart_base_seq512_model.bin \\n --world_size 8 --gpu_ranks 0 1 2 3 4 5 6 7 \\n --total_steps 1000000 --save_checkpoint_steps 100000 --report_steps 50000 \\n --learning_rate 1e-4 --batch_size 16 \\n --span_masking --span_max_length 3 \\n --embedding word_pos --tgt_embedding word_pos \\n --encoder transformer --mask fully_visible --decoder transformer \\n --target bart --tie_weights --has_lmtarget_bias
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 44 |
```
|
| 45 |
|
| 46 |
Finally, we convert the pre-trained model into Huggingface's format:
|