CODEWITHSUNDEEP
amazon-ec2neural-networktorchlstm
Adam12344
Adam12344

Reputation: 1053

Training LSTM in Torch

I'm trying to implement an LSTM model in Torch on EC2.

I don't have any experience with Torch but I found this code: http://apaszke.github.io/lstm-explained.html

I have Torch running on EC2 with the LSTM.lua script loaded.

I read through the guide but I'm not sure how to add my data and train the model. Do I just put a :forward and a :backward in a loop?

Upvotes: 2

Views: 486

Answers (1)

Manuel Lagunas
Manuel Lagunas

Reputation: 2751

You can have a look at this web page. https://github.com/torch/nn/blob/master/doc/training.md#nn.traningneuralnet.dok

input(data) and output(labels) can be Tensors so you can load your data in the Torch structure, then :forward, :backward and upgrade the model parameters.

Hope this helps.

Upvotes: 0

Related Questions