Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

enable layer group to output a sequence inside it during generation #2871

Merged
merged 7 commits into from
Aug 23, 2017
Merged

enable layer group to output a sequence inside it during generation #2871

merged 7 commits into from
Aug 23, 2017

Conversation

lcy-seso
Copy link
Contributor

@lcy-seso lcy-seso commented Jul 14, 2017 •
edited by luotao1
Loading

fixes #1797
fixes #1269

This PR mainly contains three modifications to RecurrentLayerGroup:

  1. store and return log probability for each generated word.
  2. enable recurrent layer group to return a layer's output during generating each sequence, rather than only a sequence with the highest probability.
  3. enable recurrent layer group to return a sequence layer's output during generating.

@lcy-seso lcy-seso requested a review from luotao1 July 14, 2017 09:32
@luotao1 luotao1 mentioned this pull request Aug 3, 2017
Merged
luotao1
luotao1 reviewed Aug 3, 2017
View reviewed changes
Copy link
Contributor

@luotao1 luotao1 left a comment

Choose a reason for hiding this comment

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

可以把第一个commit单独拿出来先merge了。后面几个commit可以再拆的细一点，方便review。

paddle/gserver/gradientmachines/RecurrentGradientMachine.h Outdated
* sequence in Matrix stored the entire beam search batch's forward pass
* results.
*
* @param isSeq: a flag indicating whetehr the layer to be output of the
Copy link
Contributor

Choose a reason for hiding this comment

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

whetehr->whether

Copy link
Contributor Author

Choose a reason for hiding this comment

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

done.

paddle/gserver/gradientmachines/RecurrentGradientMachine.h Outdated
*
* @param isSeq: a flag indicating whetehr the layer to be output of the
* RecurrentGradientMachine is a sequence or not
* @param outArgs: all of the the returned Arguments of the forward pass
Copy link
Contributor

Choose a reason for hiding this comment

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

the the -> the

Copy link
Contributor Author

Choose a reason for hiding this comment

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

done.

@lcy-seso lcy-seso requested review from guoshengCS and removed request for qingqing01 August 22, 2017 12:59
@lcy-seso lcy-seso merged commit 7c8acd4 into PaddlePaddle:develop Aug 23, 2017
@lcy-seso lcy-seso deleted the print_attention_weight branch August 24, 2017 03:21
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@luotao1 luotao1 luotao1 left review comments

@guoshengCS guoshengCS guoshengCS approved these changes

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
3 participants
@lcy-seso @luotao1 @guoshengCS