Open
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Fast Z-80 version has the alphabet integrated with the rest of the layer data. This leads to marginally faster code as the INFER subroutine will look for the largest output and select the character immediately after computing the value. This avoids an unnecessary divide by 4 and a second loop over the final outputs. I also moved the ReLU decision upstream a bit so we don't pointlessly divide negative outputs by 4.
All the data in one big blob makes it easier to factor out the data generation into a common routine and also raises the possibility of having a Z-80 "shell" that can load a variety of models.
There is the usual set of extra Z-80 operations added to libz80.py and a big hack that allows using "label+1" to generate the value of a label + 1.