Skip to content

加速方法? #4

Description

@ShuGao0810

博主好,训练7G+的语料时,当前代码运行效率非常低。我试了下用GPU加速训练,在cw2vec_kernals.cc的
REGISTER_KERNEL_BUILDER(Name("NegTrainCw2vec").Device(DEVICE_CPU), NegTrainCw2vecOp);后加上了

#if GOOGLE_CUDA
REGISTER_KERNEL_BUILDER(Name("NegTrainCw2vec").Device(DEVICE_GPU), NegTrainCw2vecOp);
#endif

同时在shell脚本中指定CUDA_VISIBLE_DEVICES='0',似乎不起作用,博主这边有没有加速建议?可以从哪些方面加速?

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type
    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions