You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@systemml.apache.org by "Niketan Pansare (JIRA)" <ji...@apache.org> on 2017/09/28 18:00:01 UTC
[jira] [Commented] (SYSTEMML-1500) Add missing loss layers to
Caffe2DML
[ https://issues.apache.org/jira/browse/SYSTEMML-1500?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16184560#comment-16184560 ]
Niketan Pansare commented on SYSTEMML-1500:
-------------------------------------------
- Euclidean loss added in the commit: https://github.com/apache/systemml/commit/61dcc85e48a390c1bb63ee4c42aad9a3fade7d06
> Add missing loss layers to Caffe2DML
> ------------------------------------
>
> Key: SYSTEMML-1500
> URL: https://issues.apache.org/jira/browse/SYSTEMML-1500
> Project: SystemML
> Issue Type: Sub-task
> Reporter: Niketan Pansare
>
> Multinomial Logistic Loss
> Infogain Loss - a generalization of MultinomialLogisticLossLayer.
> Softmax with Loss - computes the multinomial logistic loss of the softmax of its inputs. It’s conceptually identical to a softmax layer followed by a multinomial logistic loss layer, but provides a more numerically stable gradient.
> Sum-of-Squares / Euclidean - computes the sum of squares of differences of its two inputs, 12N∑Ni=1∥x1i−x2i∥2212N∑i=1N‖xi1−xi2‖22.
> Hinge / Margin - The hinge loss layer computes a one-vs-all hinge (L1) or squared hinge loss (L2).
> Sigmoid Cross-Entropy Loss - computes the cross-entropy (logistic) loss, often used for predicting targets interpreted as probabilities.
> Accuracy / Top-k layer - scores the output as an accuracy with respect to target – it is not actually a loss and has no backward step.
> Contrastive Loss
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)