This repository has been archived by the owner on Nov 17, 2023. It is now read-only.
R RNN API fixes and Optimizer clip gradient on NDArray #9022
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
@thirdwing : I think these changes should bring a broader usability of the RNN API. |
|
I will look into this now. Sorry for the late reply. Too busy with the thesis. |
thirdwing
approved these changes
Dec 19, 2017
|
Can you also update the code in https://github.com/dmlc/dmlc.github.io/blob/master/_posts/2017-10-11-rnn-bucket-mxnet-R.md ? |
|
Yes, just created the pull requests on dmlc.github.io and web-data repos. |
meissnereric
pushed a commit
to meissnereric/incubator-mxnet
that referenced
this pull request
Jan 2, 2018
* NDArray optimizer gradient clip and RNN shapes rework * typo bucketIter * R RNN fix typos and example
yuxiangw
pushed a commit
to yuxiangw/incubator-mxnet
that referenced
this pull request
Jan 25, 2018
* NDArray optimizer gradient clip and RNN shapes rework * typo bucketIter * R RNN fix typos and example
rahul003
pushed a commit
to rahul003/mxnet
that referenced
this pull request
Jun 4, 2018
* NDArray optimizer gradient clip and RNN shapes rework * typo bucketIter * R RNN fix typos and example
zheng-da
pushed a commit
to zheng-da/incubator-mxnet
that referenced
this pull request
Jun 28, 2018
* NDArray optimizer gradient clip and RNN shapes rework * typo bucketIter * R RNN fix typos and example
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Description
Rework of the RNN API:
seq_len X batch.size, this will ease the use of existing arrayIter on NLP models.symbol.RNN)Optimizers:
For all optimizers, gradient clipping is now performed by
mx.nd.clip, avoiding to back and forth with R array resulting in some speedup.Following comments on lack of exmaples supporting the R package, exemples have been added:
https://jeremiedb.github.io/mxnet_R_bucketing/index.html
If there's interest, idea to is add other examples, in the spirit of the gluon.mxnet.io notebooks
Checklist
Essentials