Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

TODOs #1

Open
7 of 10 tasks
vasqu opened this issue Oct 5, 2024 · 0 comments
Open
7 of 10 tasks

TODOs #1

vasqu opened this issue Oct 5, 2024 · 0 comments
Labels
WIP Work in progress

Comments

@vasqu
Copy link
Owner

vasqu commented Oct 5, 2024 •
edited
Loading

Just so i dont forget

  • check spelling
  • add block import and usage example
  • clarify attention stuff
  • low prio and possibly not worth it: make a PR or so to lucidrains repo to fix his gru implementation on padding? (+ maybe smthn more ill have to check)
  • more docs/comments in the code esp references could help
  • fix my unnecessary additional padding in gru :)
  • add initial states support (right padding only) --> makes batched generation harder
  • fix gru caching (first last hidden state is correctly saved but the sequential afterwards needs to be updated)
  • add hidden act config for conv1d
  • maybe train a very small model on tiny shakespeare or whatever

Will update based on my progress
@vasqu vasqu added the WIP Work in progress label Oct 5, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
WIP Work in progress
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
1 participant
@vasqu