Update README.md
Browse files
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/
|
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
|
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 |
```
|