RMG-NLP #1573
Merged
RMG-NLP #1573
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
added 2 commits
April 4, 2019 17:48
Use a text-generating neural network trained on a large corpus of quotes to generate completely novel quotes and print them at the end of RMG jobs.
cgrambow
added
Status: Ready for Review
|
Nominated for PR of the year. |
Codecov Report
@@ Coverage Diff @@
## master #1573 +/- ##
==========================================
- Coverage 41.87% 41.83% -0.04%
==========================================
Files 165 165
Lines 28004 28014 +10
Branches 5713 5713
==========================================
- Hits 11726 11721 -5
- Misses 15490 15504 +14
- Partials 788 789 +1
Continue to review full report at Codecov.
|
|
Weights have been updated. This is good to go. |
mliu49
approved these changes
Apr 8, 2019
5 tasks
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
Motivation or Problem
While user satisfaction might be slightly increased upon seeing that model generation completed successfully, it could be substantially elevated upon reading an intellectually satisfying and thought-provoking quotation.
Description of Changes
To address this substantial issue, I trained a neural network on a large corpus of publicly available quotations to generate completely novel "quotes" and print them at the end of RMG jobs. The neural network is contained in a newly created external package called
textgenrnn, which has been added to the RMG Anaconda cloud. The neural network weights are part of the package, so the change to RMG code is minimal.Testing
Remake the RMG Anaconda environment and run a short RMG job to test that the quote prints correctly.
Reviewer Tips
This PR is ready for review, but hold off merging for a little while. I may want to update the weights in
textgenrnnbefore this merges.