Welcome to mirror list, hosted at ThFree Co, Russian Federation.

github.com/torch/nn.git - Unnamed repository; edit this file 'description' to name the repository.
summaryrefslogtreecommitdiff
path: root/doc
diff options
context:
space:
mode:
authorSoumith Chintala <soumith@gmail.com>2017-03-23 05:39:20 +0300
committerGitHub <noreply@github.com>2017-03-23 05:39:20 +0300
commit94876e756167629476a4e59dbf415dbee5b30abd (patch)
tree37d8fe315e5c55f12deea74779b137de60f0f602 /doc
parent57d91b902c0fc963facfe59b44ffa18c0268fcc5 (diff)
parentb4a58c5c93966eb7f11587548decacd1a9b955be (diff)
Merge pull request #1167 from twitwi/patch-1
Fix possible math typo in MultiMarginCriterion doc
Diffstat (limited to 'doc')
-rw-r--r--doc/criterion.md2
1 files changed, 1 insertions, 1 deletions
diff --git a/doc/criterion.md b/doc/criterion.md
index cb2bbd0..d30d1e3 100644
--- a/doc/criterion.md
+++ b/doc/criterion.md
@@ -419,7 +419,7 @@ Optionally, you can give non-equal weighting on the classes by passing a 1D `wei
The loss function then becomes:
```lua
-loss(x, y) = sum_i(max(0, w[y] * (margin - x[y] - x[i]))^p) / x:size(1)
+loss(x, y) = sum_i(max(0, w[y] * (margin - x[y] + x[i]))^p) / x:size(1)
```
This criterion is especially useful for classification when used in conjunction with a module ending in the following output layer: