Skip to content

Conversation

@JiayiFeng
Copy link
Collaborator

@JiayiFeng JiayiFeng commented Mar 2, 2018

Add layer wrappers for save/load op.

This is a part of #8521

@panyx0718 panyx0718 requested review from panyx0718 and reyoung March 5, 2018 02:39
@JiayiFeng JiayiFeng requested a review from QiJune March 5, 2018 04:08
QiJune
QiJune previously approved these changes Mar 5, 2018
Copy link
Member

@QiJune QiJune left a comment

Choose a reason for hiding this comment

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

LGTM

args={"file_path": file_path})


def load_combine(out, file_path):
Copy link
Contributor

Choose a reason for hiding this comment

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

need better doc about the difference between save/load and save/load_combine

Copy link
Collaborator Author

@JiayiFeng JiayiFeng Mar 5, 2018

Choose a reason for hiding this comment

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

The main difference between save/load and combine save/load is that the combine save can take a list of variables and dump all of them into a single file, while normal save can only save one variable to one file.

You are right, I can make the difference clearer.

@qingqing01
Copy link
Contributor

should not be in file : python/paddle/fluid/layers/nn.py

@JiayiFeng
Copy link
Collaborator Author

@qingqing01 I have moved them from nn.py to tensor.py.

@JiayiFeng JiayiFeng merged commit 5b63568 into PaddlePaddle:develop Mar 5, 2018
@JiayiFeng JiayiFeng deleted the add_save_load_layer branch March 5, 2018 08:02
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.

4 participants