zhangyiss/mlx-examples
1
0
Fork 0
mirror of https://github.com/ml-explore/mlx-examples.git synced 2025-06-28 20:25:22 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity

nits

Browse Source
This commit is contained in:
Goekdeniz-Guelmez 2025-02-13 21:46:30 +01:00
parent 8179b99436
commit 65a49dda0e
1 changed files with 3 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
llms/mlx_lm/tuner/datasets.py
View File

@ -34,6 +34,7 @@ class GRPODataset:
The reasoning process and answer are enclosed within <think> </think> and <answer> </answer> tags, respectively, i.e., <think> reasoning process here </think><answer> answer here </answer>."""}, The reasoning process and answer are enclosed within <think> </think> and <answer> </answer> tags, respectively, i.e., <think> reasoning process here </think><answer> answer here </answer>."""},
{'role': 'user', 'content': prompt_str} {'role': 'user', 'content': prompt_str}
], ],
add_generation_prompt=True
) )
answer_tokens = tokenizer.encode(answer_str) answer_tokens = tokenizer.encode(answer_str)
else: else:
@ -307,10 +308,10 @@ def load_dataset(args, tokenizer: PreTrainedTokenizer):
else: else:
data_path = Path(args.data) data_path = Path(args.data)
if data_path.exists(): if data_path.exists():
train, valid, test = load_local_dataset(args, data_path, tokenizer, args) train, valid, test = load_local_dataset(args, data_path, tokenizer, args.config)
else: else:
print(f"Loading Hugging Face dataset {args.data}.") print(f"Loading Hugging Face dataset {args.data}.")
train, valid, test = load_hf_dataset(args, args.data, tokenizer, args) train, valid, test = load_hf_dataset(args, args.data, tokenizer, args.config)
if args.train and len(train) == 0: if args.train and len(train) == 0:
raise ValueError( raise ValueError(