Github user loachli commented on the pull request:
https://github.com/apache/spark/pull/1290#issuecomment-102955931
@avulanov I use a small fraction of mnist data (100 for training, 100 for
testing). When I use "Topology.multiLayerPerceptron(Array[Int](784,10),
false)", the accuracy is 0.68. When I use "
Topology.multiLayerPerceptron(Array[Int](784,32,10), false)", the accuracy is
0.52. That is to say, adding hidden lays might decrease the accuracy.
Could you explain the design idea of deltas in the function
"FeedForwardModel.computeGradient". I do not understand the following code :
" deltas(L) = new BDM[Double](0, 0)
deltas(L - 1) = newE
for (i <- (L - 2) to (0, -1)) {
deltas(i) = layerModels(i + 1).prevDelta(deltas(i + 1), outputs(i +
1))
}"
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
