github/ColossalAI
1
0
Fork 0
mirror of https://github.com/hpcaitech/ColossalAI.git synced 2025-07-27 13:30:14 +00:00
Code Issues Packages Projects Releases Wiki Activity

update to conform to json format

Browse Source
This commit is contained in:
Tong Li 2025-05-28 13:13:21 +08:00 committed by xysheng-colossal
parent 2dd59c0ddd
commit 2783ddecf5
2 changed files with 2 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
applications/ColossalChat/coati/distributed/README.md
View File

@ -178,7 +178,7 @@ Each data sample in the training or evaluation `.jsonl` file should follow this
| ---------------- | --------------------------------------- | ----------------- |
| `--model` | Model path or identifier | `/path/to/model` |
| `--dataset` | Path to training `.jsonl` | `/path/to/train_data.jsonl` |
| `--eval-dataset` | JSON of task\:eval\_dataset\_path pairs | `{'eval_1':'/path/to/eval_1.jsonl'}` |
| `--eval-dataset` | JSON of task\:eval\_dataset\_path pairs | `{"eval_1":"/path/to/eval_1.jsonl"}` |
| `--project` | Project name | `Project1` |
| `--num-episodes` | Number of training episodes | `1` |

2
applications/ColossalChat/rl_example.py
View File

@ -23,7 +23,7 @@ if __name__ == "__main__":
default=None,
help="Evaluation dataset for each task, please use json format to specify the dataset for each task. \
For example: {'task1':'data_eval_task1.jsonl', 'task2':'data_eval_task2.jsonl'}, the jsonl file should be in the same format as the training dataset. \
The key is the task name, and the value is the path to the jsonl file",
The key is the task name, and the value is the path to the jsonl file, please replace sinple quotes with double quotes to conform to json format.",
)
parser.add_argument("-p", "--project", type=str, default="GRPO", help="Project name.")
parser.add_argument("-e", "--num-episodes", type=int, default=1, help="Number of episodes to train.")