change data to sequences processing #3
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
All the lines having smth like data[i*seq_length:(i+1)*seq_length] actually get 1/seq_length of the training data, as they convert data to disjoint sequences of seq_length length. Proposing a small enhancement - to gather all char sequences from the data, with step 1. That will slower the training, but will correct the monstrous loss of data.
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Outdated suggestions cannot be applied.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.
All the lines having smth like data[i*seq_length:(i+1)*seq_length] actually get 1/seq_length of the training data, as they convert data to disjoint sequences of seq_length length. Proposing a small enhancement - to gather all char sequences from the data, with step 1. That will slower the training, but will correct the monstrous loss of data.