Skip to content

Conversation

@QiJune
Copy link
Member

@QiJune QiJune commented Jun 11, 2018

Try to fix #11320



def memory_optimize(input_program, print_log=False, level=0):
def memory_optimize(input_program, skip_opt_set=None, print_log=False, level=0):
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

  1. Need to add comments for the new arguments.
  2. Need unit test.

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

直接修改本地安装环境中对应文件,进行测试,获取的loss值依旧异常

Copy link
Member Author

@QiJune QiJune Jun 12, 2018

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@zchengquan 你是从源码编译安装吗?
不是很确定直接修改本地安装环境中的对应文件,能否使得这个skip生效。

如果是从源码编译安装的话,在修改对应python文件后,请重新执行make install

Copy link
Contributor

@dzhwinter dzhwinter left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Great PR, had help with @dyning fix this issue. I will do polish the api and add more tests in #11462

@dzhwinter dzhwinter merged commit 34ac0eb into PaddlePaddle:develop Jun 15, 2018
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

启动内存优化后,多个loss训练,获取loss值有异常

4 participants