AI软件下载
有趣网站推荐及实用软件下载

RuntimeError: Distributed package doesn't have NCCL built in

在调试python应用MAGI-1的时候发生报错,部分错误信息如下:

 File "D:\python\MAGI-1\inference\infra\distributed\dist_utils.py", line 42, in dist_init
    torch.distributed.init_process_group(
  File "D:\python\MAGI-1\py310\lib\site-packages\torch\distributed\c10d_logger.py", line 79, in wrapper
    return func(*args, **kwargs)
  File "D:\python\MAGI-1\py310\lib\site-packages\torch\distributed\c10d_logger.py", line 93, in wrapper
    func_return = func(*args, **kwargs)
  File "D:\python\MAGI-1\py310\lib\site-packages\torch\distributed\distributed_c10d.py", line 1368, in init_process_group
    default_pg, _ = _new_process_group_helper(
  File "D:\python\MAGI-1\py310\lib\site-packages\torch\distributed\distributed_c10d.py", line 1573, in _new_process_group_helper
    raise RuntimeError("Distributed package doesn't have NCCL built in")
RuntimeError: Distributed package doesn't have NCCL built in

解决方法将NCCL改为gloo

inference\infra\distributed\dist_utils.py第43行

backend=config.engine_config.distributed_backend,

改为

backend='gloo',


AI软件用不了?2元爽玩4090: 立即体验>>

软件催更及1对1人工答疑支持: https://nuowa.net/1806
赞(0) 打赏
软件无法使用?点击查看常见问题说明>>

最近更新

觉得文章对你有帮助就打赏一下作者

非常感谢你的打赏,我将有更多的动力继续提供优质内容,让我们一起创建更加美好的世界!

支付宝扫一扫

微信扫一扫