Update from original #1
Merged
Conversation
This file contains hidden or 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
Fix unitnorm to be a class, like other constraints
New pull request -- much cleaner
Fix parenthesis typo in examples.md
Proper handling of output values in Masking layer
Updated adam solver to v8 of paper
Added Poisson Loss as an objective
urlretrieve will blindly swallow any 4xx and 5xx responses and then save the html error response in the local file. This is probably exactly what we don't want, because not only will the program crash if there is a network hiccup when the error file cannot be opened, but it will continue to do so when rerun until the corrupt cached file is found and manually removed. Luckily, urlretrieve is just a thin wrapper around FancyURLopener, so we can make our own thin wrapper that throws an exception instead of caching the wrong file. Tested to be working as before when running cached and uncached datasets, and also verified to fail loudly when asked to fetch http://httpstat.us/500
There was some confusion around whether you could mask individual timesteps.
Should downsample on `steps`.
For all other questions, the full vocab is in the stories and the queries
+ reduce function disappeared (requires import from functools) + tarfiles and encodings - decoding bytes to ASCII at line level
Fixes and full results for bAbi RNN example
data_utils: add error handling on url fetches
Add documentation for time distributed sample weighting
…into nehz-nehz-MaxPooling1D-fix
Fix conv layers loading for model_from_config
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
20 participants
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Outdated suggestions cannot be applied.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.
No description provided.