Skip to content

Long-short term memory #1380

@saska

Description

@saska

Hey everyone,

I'd like to look into implementing LSTM in this library, so probably in the neural_network directory. I've got a repo I'd like to use as a base, and it's under the same license - does anyone how this works? Should I include my little copyright in there since the MIT license says it needs to be present (the other stuff required is already found in the license of TheAlgorithms/Python)? (EDIT: I.e. can I do a new version that's free of my copyright or is self-plagiarism still plagiarism or how does this stuff actually work?)

Does this make sense? Other than implementing a properly working gradient checker and refactoring my code to a single file to comply with the folder structure, is there anything I need to do? Where should I add my tests (I see some of the other folders already have a tests directory, should neural networks have one as well)? I also have some pre-written explanatory stuff for it (Algorithms-Explanation), although I definitely don't think I can calculate the time complexity of this thing...

Metadata

Metadata

Assignees

No one assigned

    Labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions