huggingface config_error ai_generated true

值错误：当设备明确设置为 0 时，无法在 pipeline 中使用 device_map='auto'。请设置 device_map=None 或移除 device 参数。

ValueError: Cannot use device_map='auto' with pipeline when device is explicitly set to 0. Please set device_map=None or remove device argument.

ID: huggingface/pipeline-device-map-conflict

其他格式: JSON · Markdown 中文 · English
92%修复率
90%置信度
1证据数
2024-03-22首次发现

版本兼容性

版本状态引入弃用备注
transformers>=4.30.0 active — — —
accelerate>=0.20.0 active — — —

根因分析

当同时指定 `device` 和 `device_map` 时，pipeline API 发生冲突；`device_map='auto'` 要求模型使用 Accelerate 的设备映射加载，这与固定的设备索引不兼容。

English

The pipeline API conflicts when both `device` and `device_map` are specified; `device_map='auto'` requires the model to be loaded with Accelerate's device mapping, which is incompatible with a fixed device index.

generic

官方文档

https://huggingface.co/docs/transformers/v4.30.0/en/main_classes/pipelines#pipeline-device-map

解决方案

  1. 移除 `device` 参数，仅使用 `device_map='auto'`：pipe = pipeline('text-generation', model='model-name', device_map='auto')。这使 Accelerate 处理设备放置。
  2. 移除 `device_map` 并显式使用 `device=0`：pipe = pipeline('text-generation', model='model-name', device=0)。这强制模型使用 GPU 0。

无效尝试

常见但无效的做法:

  1. Set both `device=0` and `device_map='auto'` and expect the pipeline to resolve the conflict 100% 失败

    The pipeline raises a ValueError immediately because the two arguments are mutually exclusive.

  2. Use `device_map='sequential'` instead of 'auto' 90% 失败

    The same conflict applies; any non-None device_map with an explicit device argument will raise an error.

  3. Set `device=-1` to use CPU and keep device_map='auto' 80% 失败

    device_map='auto' requires at least one GPU to be available; setting device=-1 will cause a separate error about no GPU found.