Unable to Load New Training Set After Completing One Set of n Epochs

[GoogleCodeExporter]

What steps will reproduce the problem?
1. Load training data and configure network.  Train.
2. After n epochs are exhausted, the Training Set box is grayed out, so one is 
unable to load new training data.

What is the expected output? What do you see instead?
One would expect to be able to load new training data to continue training the 
network.

What version of the product are you using? On what operating system?
Release before 2012/09/10.  Windows 7.  nVidia GTX 550 Ti.

Please provide any additional information below.
Workaround: Export the network to disk.  Hit stop to clear the network.  Import 
new training data.  Import network from disk.  Continue training with new 
dataset.

Original issue reported on code.google.com by jo.j...@gmail.com on 11 Sep 2012 at 3:38

[GoogleCodeExporter]

This was by design, though I think I'll fix this to work as expected once I'm 
done implementing dropout.

Original comment by pospes...@gmail.com on 11 Sep 2012 at 3:44

• Changed state: Accepted
• Added labels: Type-Task
• Removed labels: Type-Defect

[GoogleCodeExporter]

Dropout is a far more interesting feature.  I await anxiously.

Original comment by jo.j...@gmail.com on 11 Sep 2012 at 4:21

[GoogleCodeExporter]

I'll fix this with the next release, which will be mostly GUI fixes.

Original comment by pospes...@gmail.com on 9 Oct 2012 at 2:00

[GoogleCodeExporter]

Fixed in latest published build (r38)

Original comment by pospes...@gmail.com on 14 Oct 2012 at 11:07

• Changed state: Fixed