00:34 < zoq> mlPackApprentice: If the responses parameter has only one slice the prediction will predict the next hour, in this case you might want to set single = true in the RNN constructor.

00:34 < zoq> mlPackApprentice: What output layer do you use?

07:12 < wannabeOG> Hi I am Saihimal Allu. I am a junior majoring at IIT Roorkee. I have been involved in deep learning and it's applications in Computer Vision since the last 2 years. I would love to contribute to this organization. I have gone through the links which you guys had posted on official page of ml-pack regarding some of the basics of C++. Are there any other pointers that I need to know in addition to them?

09:04 < zoq> wannabeOG: Hello there, do you mean mlpack.org/involved.html and https://www.mlpack.org/gsoc.html?

09:48 < jenkins-mlpack2> Project docker mlpack weekly build build #23: STILL UNSTABLE in 11 hr: http://ci.mlpack.org/job/docker%20mlpack%20weekly%20build/23/

12:06 < jenkins-mlpack2> Project docker mlpack nightly build build #153: STILL UNSTABLE in 6 hr 52 min: http://ci.mlpack.org/job/docker%20mlpack%20nightly%20build/153/

14:03 < prajjwal> hello i am new to mlpack and looking forward to contribute to it. Can anyone tell me where and how to start? Thankyou!!!

15:13 < mlPackApprentice> @zoq Hello, to continue on this question "00:34 < zoq> mlPackApprentice: What output layer do you use?", our RNN configuration is as follows:

15:15 < mlPackApprentice> Identity Layer

15:15 < mlPackApprentice> Linear Layer

15:15 < mlPackApprentice> FastLSTM

15:15 < mlPackApprentice> LogSoftMax

15:16 < mlPackApprentice> in that order

15:19 < mlPackApprentice> So a LogSoftMax layer would be the output layer. The FastLSTM has 16 Hidden nodes and feeds to 5 Output nodes. Those 5 correspond to the first 5 input nodes.