[
https://issues.apache.org/jira/browse/SPARK-5575?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14996706#comment-14996706
]
Narine Kokhlikyan commented on SPARK-5575:
------------------------------------------
Hi [~avulanov] ,
I was trying out the current implementation of ANN and have one question about
it.
Usually, when I run neuronal network with other tools such as R, I can
additionally see information about: e.g. Error, Reached Threshold and Steps.
Can I also somehow get such information from Spark ANN ? Maybe it is already
there, I couldn't find it.
I looked through the implementations of GradientDecent and LBFGS and it seems
that the optimizer.optimize doesn't return values about the error, number of
iterations, etc.
I might be wrong here, still investigating it, but, I'd be happy to hear from
you regarding this.
Thanks,
Narine
> Artificial neural networks for MLlib deep learning
> --------------------------------------------------
>
> Key: SPARK-5575
> URL: https://issues.apache.org/jira/browse/SPARK-5575
> Project: Spark
> Issue Type: Umbrella
> Components: MLlib
> Affects Versions: 1.2.0
> Reporter: Alexander Ulanov
>
> Goal: Implement various types of artificial neural networks
> Motivation: deep learning trend
> Requirements:
> 1) Basic abstractions such as Neuron, Layer, Error, Regularization, Forward
> and Backpropagation etc. should be implemented as traits or interfaces, so
> they can be easily extended or reused
> 2) Implement complex abstractions, such as feed forward and recurrent networks
> 3) Implement multilayer perceptron (MLP), convolutional networks (LeNet),
> autoencoder (sparse and denoising), stacked autoencoder, restricted
> boltzmann machines (RBM), deep belief networks (DBN) etc.
> 4) Implement or reuse supporting constucts, such as classifiers, normalizers,
> poolers, etc.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
