Giters

Andy-P / RecurrentNN.jl

Deep RNN, LSTM, GRU, GF-RNN, and GF-LSTMs in Julia

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Improving the design

Andy-P opened this issue · comments

commented

Starting a thread to improve the overall design. I think this approach of using graph.jl and solver.jl to create arbitrary deep learning models is really interesting and could evolve into something really useful to the community.

commented

Just putting together a list of possible improvement/ideas in no particular order

  1. The input to the LSTM and RNN models shouldn't have to be a NNMatrix
  2. The graph.jl and solver.jl might better as a separate library/dependancy (NNGraph.jl?)
  3. The Abstract type Model is too generic a name. Change to NNModel?

Will add more as they surface.