uer commited on
Commit
b60f206
1 Parent(s): eb5d751

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +2 -2
README.md CHANGED
@@ -57,7 +57,7 @@ python3 preprocess.py --corpus_path corpora/couplet.txt \
57
  ```
58
  python3 pretrain.py --dataset_path couplet_dataset.pt \
59
  --vocab_path models/google_zh_vocab.txt \
60
- --output_model_path models/couplet_gpt2_base_model.bin \
61
  --config_path models/gpt2/config.json \
62
  --world_size 8 --gpu_ranks 0 1 2 3 4 5 6 7 \
63
  --total_steps 25000 --save_checkpoint_steps 5000 --report_steps 1000 \
@@ -71,7 +71,7 @@ python3 pretrain.py --dataset_path couplet_dataset.pt \
71
 
72
  Finally, we convert the pre-trained model into Huggingface's format:
73
  ```
74
- python3 scripts/convert_gpt2_from_uer_to_huggingface.py --input_model_path couplet_gpt2_base_model.bin-25000 \
75
  --output_model_path pytorch_model.bin \
76
  --layers_num 12
77
  ```
 
57
  ```
58
  python3 pretrain.py --dataset_path couplet_dataset.pt \
59
  --vocab_path models/google_zh_vocab.txt \
60
+ --output_model_path models/couplet_gpt2_model.bin \
61
  --config_path models/gpt2/config.json \
62
  --world_size 8 --gpu_ranks 0 1 2 3 4 5 6 7 \
63
  --total_steps 25000 --save_checkpoint_steps 5000 --report_steps 1000 \
 
71
 
72
  Finally, we convert the pre-trained model into Huggingface's format:
73
  ```
74
+ python3 scripts/convert_gpt2_from_uer_to_huggingface.py --input_model_path couplet_gpt2_model.bin-25000 \
75
  --output_model_path pytorch_model.bin \
76
  --layers_num 12
77
  ```