1) add weight init method from RNNPlus to RNN 2) implement RNNs + Attention

timeseriesAI / tsai

Time series Timeseries Deep Learning Machine Learning Python Pytorch fastai | State-of-the-art Deep Learning library for Time Series and Sequences in Pytorch / fastai

https://timeseriesai.github.io/tsai/

Apache License 2.0

5.21k stars 651 forks source link

1) add weight init method from RNNPlus to RNN 2) implement RNNs + Attention #616

Closed yangtzech closed 1 year ago

yangtzech commented 1 year ago

Sorry that I put two unrelated commits in one pull request. 1) the first commit is to implement the weights initialization method which has been implemented in the RNNPlus 2) I am recently trying to use RNN+Attention, and I see #595 . I think it would be useful to make use of the utilities those tsai provide and therefore, I made this implementation. I just put an RNN layer before the encoder layer, so please tell me if I misunderstood it.

review-notebook-app[bot] commented 1 year ago

Check out this pull request on

See visual diffs & provide feedback on Jupyter Notebooks.

Powered by ReviewNB

yangtzech commented 1 year ago

I think I have to modify it again