Skip to content

Possible to specify hidden dimension separately from embedding dimension? #109

Description

@AngledLuffa

For example, with a PyTorch LSTM, one can do

LSTM(input_size, hidden_size, ...)

Is there a way to seperate the input_size from the hidden_size in this model? Or is the output size always the same as the input?

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions