Commits

Breadcrumbs

History for

onmaster

User selector

All users

Datepicker

All time

Commit History

Commits on Nov 14, 2016

• PairwiseDistance: Default using Euclidean distance.

  

  Sagar_M_Waghmare

  authored and

  Sagar_M_Waghmare

  committed
  6cbe21b

Commits on Feb 9, 2016

• nn.clearState

  szagoruyko

  committed
  b4ebdf2

Commits on Dec 15, 2015

• Merge pull request torch#532 from xwgeng/master

  Show description for 9f635fb
  soumith

  committed
  9f635fb

• the statements are redundant

  xwgeng

  committed
  a939077

Commits on Dec 11, 2015

• PairwiseDistance: output and gradInput initialized on use and not in constructor (allows proper sanitization).

  mys007

  committed
  fd9cd3c

Commits on Apr 28, 2015

• Make type() truly recursive.

  Show description for d0da63e
  dominikgrewe

  committed
  d0da63e

Commits on Nov 18, 2013

• fixed PairwiseDistance type() function bug. Overloaded type now returns self.

  jonathantompson

  committed
  4734d24

• Fixed a bug in PairwiseDistance where the gradInput table isn't converted when Module.type function is called (this bug has always existed and is not due to the recent changes).

  jonathantompson

  committed
  f87f741

Commits on Oct 21, 2013

• going back on Clement's suggestion. It was a good idea, but we're needlessly loosing performance in the 1D case with the input clone.

  jonathantompson

  committed
  3ce35f0

Commits on Oct 20, 2013

• fixed a bug in Pairwise distance when the output Lp norm is zero (which results in a divide by zero issue). Rewrote PairwiseDistance following Clement's suggestion to only have one codepath. Fixed …

  Show description for 7286784
  jonathantompson

  committed
  7286784

Commits on Oct 19, 2013

• it seems like torch.abs() doesn't have a cuda implementation, so the previous commit would fail when PairwiseDistance():cuda() was called.

  jonathantompson

  committed
  e1fbc0c

Commits on Oct 18, 2013

• Fixed the bprop in PairwiseDistance for pnorms other than one. The gradInput has always been wrong it seems; the sign of the gradient is correct but the magnitude was wrong. I also added a test in …

  Show description for 4be845e
  jonathantompson

  committed
  4be845e

Commits on Oct 17, 2013

• Fixed PairwiseDistance for odd Lp norms

  jonathantompson

  committed
  d4792ac

Commits on Oct 16, 2013

• init self.output with a single element tensor, not an empty tensor

  koraykv

  committed
  b8f17d4

Commits on Sep 4, 2013

• Added minibatch support to MarginRankingCriterion and PairwiseDistance.

  vladmnih

  committed
  4b3a22b

Commits on Jan 25, 2012

• initial revamp of torch7 tree

  andresy

  committed
  4df3893

Loading