search

cbovar / ConvNetSharp

Deep Learning in C#
MIT License
467 stars 109 forks source link

Distributed learning #57

Open glaporte opened 7 years ago

glaporte commented 7 years ago

Hello, is there a solution for distributed learning ? I think the most obvious is to create a small server then send samples to convnetsharp?

Support of RNN is coming soon?

Thank you:)

cbovar commented 7 years ago

Hi, There is not solution for distributed learning so far. Yes, I guess you could train different nodes with different samples set and share the weights from time to time. I am not sure of the strategy to share weights among nodes (average them ?).

Currently I am fixing/improving ConvNet.Fluent part. My short term objective is to have a working convolutional autoencoder. Once this will be done I will continue my work on LSTM

glaporte commented 7 years ago

Hi,

Thank you for your information :) I am waiting for LSTM 👍

Keep the good work!