qlora微调Qwen2.5-Omni模型报错ValueError: Processor was not found, please check and update your processor config.

[leidaoyu]

Reminder

• I have read the above rules and searched the existing issues.

System Info

llamafactory version: 0.9.3.dev0
Python version: 3.11.11
PyTorch version: 2.5.1+cu124
Transformers version: 4.52.0.dev0

Reproduction

"""
Traceback (most recent call last):
File "C:\Users\76425.conda\envs\torch\Lib\site-packages\multiprocess\pool.py", line 125, in worker
result = (True, func(*args, **kwds))
^^^^^^^^^^^^^^^^^^^
File "C:\Users\76425.conda\envs\torch\Lib\site-packages\datasets\utils\py_utils.py", line 680, in _write_generator_to_queue
for i, result in enumerate(func(**kwargs)):
File "C:\Users\76425.conda\envs\torch\Lib\site-packages\datasets\arrow_dataset.py", line 3516, in _map_single
for i, batch in iter_outputs(shard_iterable):
File "C:\Users\76425.conda\envs\torch\Lib\site-packages\datasets\arrow_dataset.py", line 3466, in iter_outputs
yield i, apply_function(example, i, offset=offset)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\76425.conda\envs\torch\Lib\site-packages\datasets\arrow_dataset.py", line 3389, in apply_function
processed_inputs = function(*fn_args, *additional_args, **fn_kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\76425\Desktop\LLaMA-Factory\src\llamafactory\data\processor\supervised.py", line 99, in preprocess_dataset
input_ids, labels = self._encode_data_example(
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\76425\Desktop\LLaMA-Factory\src\llamafactory\data\processor\supervised.py", line 43, in _encode_data_example
messages = self.template.mm_plugin.process_messages(prompt + response, images, videos, audios, self.processor)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\76425\Desktop\LLaMA-Factory\src\llamafactory\data\mm_plugin.py", line 1584, in process_messages
self._validate_input(processor, images, videos, audios)
raise ValueError("Processor was not found, please check and update your processor config.")
ValueError: Processor was not found, please check and update your processor config.
"""

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
File "", line 198, in _run_module_as_main
File "", line 88, in run_code
File "C:\Users\76425.conda\envs\torch\Scripts\llamafactory-cli.exe_main.py", line 7, in
File "C:\Users\76425\Desktop\LLaMA-Factory\src\llamafactory\cli.py", line 115, in main
COMMAND_MAPcommand
File "C:\Users\76425\Desktop\LLaMA-Factory\src\llamafactory\train\tuner.py", line 107, in run_exp
_training_function(config={"args": args, "callbacks": callbacks})
File "C:\Users\76425\Desktop\LLaMA-Factory\src\llamafactory\train\tuner.py", line 69, in _training_function
run_sft(model_args, data_args, training_args, finetuning_args, generating_args, callbacks)
File "C:\Users\76425\Desktop\LLaMA-Factory\src\llamafactory\train\sft\workflow.py", line 51, in run_sft
dataset_module = get_dataset(template, model_args, data_args, training_args, stage="sft", **tokenizer_module)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\76425\Desktop\LLaMA-Factory\src\llamafactory\data\loader.py", line 310, in get_dataset
dataset = _get_preprocessed_dataset(
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\76425\Desktop\LLaMA-Factory\src\llamafactory\data\loader.py", line 256, in _get_preprocessed_dataset
dataset = dataset.map(
^^^^^^^^^^^^
File "C:\Users\76425.conda\envs\torch\Lib\site-packages\datasets\arrow_dataset.py", line 557, in wrapper
out: Union["Dataset", "DatasetDict"] = func(self, *args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\76425.conda\envs\torch\Lib\site-packages\datasets\arrow_dataset.py", line 3166, in map
for rank, done, content in iflatmap_unordered(
File "C:\Users\76425.conda\envs\torch\Lib\site-packages\datasets\utils\py_utils.py", line 720, in iflatmap_unordered
File "C:\Users\76425.conda\envs\torch\Lib\site-packages\datasets\utils\py_utils.py", line 720, in
[async_result.get(timeout=0.05) for async_result in async_results]
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\76425.conda\envs\torch\Lib\site-packages\multiprocess\pool.py", line 774, in get
raise self._value
ValueError: Processor was not found, please check and update your processor config.

Others

按照transformers的说明pip install git+https://github.com/huggingface/[email protected]还是报错

[Kuangdd01]

pip install git+https://github.com/Kuangdd01/transformers.git@qwen25omni 暂时先用这个试一下

[leidaoyu]

pip install git+https://github.com/Kuangdd01/transformers.git@qwen25omni 暂时先用这个试一下

非常感谢！这个错解了，但是又出现一个新的错，方便看看是什么原因吗？
我的训练集json格式是：
{
"id": "example_1",
"conversations": [
{
"role": "system",
"content": "你是一个可以理解中英文语音的助手，请把输入音频转换成文本"
},
{
"role": "user",
"content": "<|AUDIO|audio_path<|AUDIO|>"
},
{
"role": "assistant",
"content": "asr结果"
}
]
}

[Kuangdd01]

参考data/mllm_audio_demo.jdon组织一下数据格式

[leidaoyu]

参考data/mllm_audio_demo.jdon组织一下数据格式

我把数据处理成了如下格式，但是仍然报错
{
"messages": [
{
"role": "system",
"content": "你是一个可以理解中英文语音的助手，请把输入音频转换成文本"
},
{
"role": "user",
"content": ""
},
{
"role": "assistant",
"content": "asr_result"
}
],
"audios": [
"audio.wav"
]
}