diff options
| author | Akhil Ketkar <akhilketkar@gmail.com> | 2016-05-01 07:18:51 +0300 |
|---|---|---|
| committer | Akhil Ketkar <akhilketkar@gmail.com> | 2016-05-01 07:18:51 +0300 |
| commit | 587d1cb8f9b4744b4a6259b6fcb4f21301e4438e (patch) | |
| tree | 69663efd5821b07c323124e70b314a65d13f4e7e | |
| parent | 954818048c089d6c7a0f13b8e07858c9a585c7d2 (diff) | |
MultiMarginCriterion description has a typo
```lua
loss(x, y) = sum_i(max(0, (margin - x[y] + x[i]))^p) / x:size(1)
```
| -rw-r--r-- | doc/criterion.md | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/doc/criterion.md b/doc/criterion.md index 78efae0..6e25f72 100644 --- a/doc/criterion.md +++ b/doc/criterion.md @@ -403,7 +403,7 @@ criterion = nn.MultiMarginCriterion(p, [weights], [margin]) Creates a criterion that optimizes a multi-class classification hinge loss (margin-based loss) between input `x` (a `Tensor` of dimension 1) and output `y` (which is a target class index, `1` <= `y` <= `x:size(1)`): ```lua -loss(x, y) = sum_i(max(0, (margin - x[y] - x[i]))^p) / x:size(1) +loss(x, y) = sum_i(max(0, (margin - x[y] + x[i]))^p) / x:size(1) ``` where `i == 1` to `x:size(1)` and `i ~= y`. |