Skip to content

我在运行您的代码时遇到了Internal: Blas GEMM launch failed 这样的情况 #2

Description

@rongmux

您好,我在尝试运行您的代码的时候,jupyter 出现了这样的错误:出现于 #开始训练 那里
InternalError: 2 root error(s) found.
(0) Internal: Blas GEMM launch failed : a.shape=(64, 32), b.shape=(32, 32), m=64, n=32, k=32
[[{{node bidirectional/while/MatMul_4}}]]
[[metrics/acc/Identity/_177]]
(1) Internal: Blas GEMM launch failed : a.shape=(64, 32), b.shape=(32, 32), m=64, n=32, k=32
[[{{node bidirectional/while/MatMul_4}}]]
0 successful operations.
0 derived errors ignored.
我的环境是tensorflow1.14 cuda10.0 cudnn是7.6 GPU是笔记本的3070,显存8G
我网上搜了很多关于这个报错的解决办法,网上说是显存不够,我修改了输入的embedding矩阵大小,benchsize大小,都解决不了,请问您有遇到过这个问题吗?您当时的GPU是什么啊?如果有遇到过,能请教一下您是怎么解决的吗?
谢谢,期待您的回复!

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions