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.

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

Add Avalon.jl MLP Model #8

Open
wants to merge 9 commits into
base: test
Choose a base branch
from
Open

Add Avalon.jl MLP Model #8

wants to merge 9 commits into from

Conversation

trevor-yu-087
Copy link
Collaborator

No description provided.

@trevor-yu-087 trevor-yu-087 self-assigned this Oct 1, 2022
@trevor-yu-087 trevor-yu-087 linked an issue Oct 1, 2022 that may be closed by this pull request
@trevor-yu-087 trevor-yu-087 marked this pull request as ready for review October 1, 2022 14:55
@trevor-yu-087 trevor-yu-087 changed the title Add avalon-mlp.jl file Add Avalon.jl MLP Model Oct 1, 2022
@trevor-yu-087
Copy link
Collaborator Author

Closes #5

@trevor-yu-087
Copy link
Collaborator Author

Notebook ready for review:
Known issues: Learning rate is 1e-3 instead of 1e-4. I think I might change the spec to 1e-3, it's more reasonable for getting better loss on this smaller model.

Avalon.jl doesn't work in its current state so relevant primitives are copied over from repo.

@trevor-yu-087 trevor-yu-087 changed the base branch from main to test October 30, 2022 01:37
@trevor-yu-087
Copy link
Collaborator Author

Script finished, ready for review

Copy link
Contributor

@UrbanPistek UrbanPistek left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Not really familar with Julia but in general my main comments are around just structure and grouping things together to be a bit more clear what everything is doing.

For example the function Linear, Net struct and then the creation of the net object - I assume these are all for initializing the model structure, I think simply grouping them together, adding a bit more comments or even making a class for the model itself just makes it clear what these are doing, the others are pretty self explainitory.

The other example would be creating a train function instead of just having it all in main().

Nothing major but these small structure changes I think would help with understading what each components is doing especially if you are unfamilar woth Julia.

I'll leave it up to our discrestion whether you want to make the changes here or just keep it in mind going forward.

@UrbanPistek
Copy link
Contributor

Another side note, again I am not familar with the Julia ecosystem but what about enviroments and dependancies? At least there should be something sepcifying dependancies and the version used here.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Projects
None yet
Development

Successfully merging this pull request may close these issues.

Avalon - MLP
2 participants