LoRA Qwen2 Error: non-default argument 'hidden_size' follows default argument

[madroidmaq]

I have encountered the following error message when fine-tuning a Qwen1.5 model with the latest mlx_lm (version 0.0.10), which was fine-tuned on version 0.0.9. The error message is as follows:
 

python -m mlx_lm.lora --model mlx-community/Qwen1.5-1.8B-Chat-4bit \
               --data /Users/madroid/Desktop/mlx-examples/lora/data/chat-im \
               --adapter-file qwen_adapters.npz \
               --lora-layers 16 \
               --batch-size 32 \
               --train \
               --test \
               --prompt "hello" \
               --iters 800
Loading pretrained model
Fetching 8 files: 100%|████████████████████████████████████████████████████████████████████████████| 8/8 [00:00<00:00, 92948.57it/s]
Traceback (most recent call last):
  File "<frozen runpy>", line 198, in _run_module_as_main
  File "<frozen runpy>", line 88, in _run_code
  File "/opt/homebrew/lib/python3.11/site-packages/mlx_lm/lora.py", line 171, in <module>
    model, tokenizer = load(args.model)
                       ^^^^^^^^^^^^^^^^
  File "/opt/homebrew/lib/python3.11/site-packages/mlx_lm/utils.py", line 278, in load
    model = load_model(model_path)
            ^^^^^^^^^^^^^^^^^^^^^^
  File "/opt/homebrew/lib/python3.11/site-packages/mlx_lm/utils.py", line 221, in load_model
    model_class, model_args_class = _get_classes(config=config)
                                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/opt/homebrew/lib/python3.11/site-packages/mlx_lm/utils.py", line 46, in _get_classes
    arch = importlib.import_module(f"mlx_lm.models.{model_type}")
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/opt/homebrew/Cellar/python@3.11/3.11.7_1/Frameworks/Python.framework/Versions/3.11/lib/python3.11/importlib/__init__.py", line 126, in import_module
    return _bootstrap._gcd_import(name[level:], package, level)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "<frozen importlib._bootstrap>", line 1204, in _gcd_import
  File "<frozen importlib._bootstrap>", line 1176, in _find_and_load
  File "<frozen importlib._bootstrap>", line 1147, in _find_and_load_unlocked
  File "<frozen importlib._bootstrap>", line 690, in _load_unlocked
  File "<frozen importlib._bootstrap_external>", line 940, in exec_module
  File "<frozen importlib._bootstrap>", line 241, in _call_with_frames_removed
  File "/opt/homebrew/lib/python3.11/site-packages/mlx_lm/models/qwen2.py", line 10, in <module>
    @dataclass
     ^^^^^^^^^
  File "/opt/homebrew/Cellar/python@3.11/3.11.7_1/Frameworks/Python.framework/Versions/3.11/lib/python3.11/dataclasses.py", line 1230, in dataclass
    return wrap(cls)
           ^^^^^^^^^
  File "/opt/homebrew/Cellar/python@3.11/3.11.7_1/Frameworks/Python.framework/Versions/3.11/lib/python3.11/dataclasses.py", line 1220, in wrap
    return _process_class(cls, init, repr, eq, order, unsafe_hash,
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/opt/homebrew/Cellar/python@3.11/3.11.7_1/Frameworks/Python.framework/Versions/3.11/lib/python3.11/dataclasses.py", line 1027, in _process_class
    _init_fn(all_init_fields,
  File "/opt/homebrew/Cellar/python@3.11/3.11.7_1/Frameworks/Python.framework/Versions/3.11/lib/python3.11/dataclasses.py", line 545, in _init_fn
    raise TypeError(f'non-default argument {f.name!r} '
TypeError: non-default argument 'hidden_size' follows default argument

[awni]

Oops, I probably broke that in #426. I will send a fix shortly.

[katopz]

@awni Not sure this released? It has been fixed in 0.0.11 am I right?

pip install mlx-lm=="0.0.11"
ERROR: Could not find a version that satisfies the requirement mlx-lm==0.0.11 (from versions: 0.0.1, 0.0.2, 0.0.3, 0.0.5, 0.0.6, 0.0.7, 0.0.8, 0.0.9.dev0, 0.0.9, 0.0.10)
ERROR: No matching distribution found for mlx-lm==0.0.11

Thanks

[awni]

Sorry I didn't push a release to pypi yet! I will do so within the next day or two!

[awni]

Should be fixed in the latest mlx lm.