frozenleaves/pytorch
1
0
Fork 0
mirror of https://github.com/pytorch/pytorch.git synced 2025-10-21 05:34:18 +08:00
Code Issues Packages Projects Releases Wiki Activity

ENH Adds label_smoothing to cross entropy loss (#63122)

Browse Source 
Summary:
Fixes https://github.com/pytorch/pytorch/issues/7455

Partially resolves pytorch/vision#4281

Pull Request resolved: https://github.com/pytorch/pytorch/pull/63122

Reviewed By: iramazanli

Differential Revision: D30586076

Pulled By: jbschlosser

fbshipit-source-id: 06afc3aa1f8b9edb07fe9ed68c58968ad1926924
This commit is contained in:
Thomas J. Fan
2021-08-29 23:31:42 -07:00
committed by Facebook GitHub Bot
parent 8af1407eab
commit d3bcba5f85
14 changed files with 412 additions and 25 deletions
Download Patch File Download Diff File Expand all files Collapse all files

25
test/cpp/api/modules.cpp
View File

@ -2315,6 +2315,31 @@ TEST_F(ModulesTest, CrossEntropyLoss) {
ASSERT_TRUE(
CrossEntropyLoss(CrossEntropyLossOptions().ignore_index(-100).reduction(torch::kMean))
->forward(input, target).allclose(expected, 1e-04));
// label smoothing with class indices
loss = CrossEntropyLoss(CrossEntropyLossOptions().label_smoothing(0.15).reduction(torch::kMean));
input = torch::tensor({{3., 1.}, {1., 2.}}, torch::dtype(torch::kFloat).requires_grad(true));
target = torch::tensor({0, 1}, torch::kLong);
output = loss->forward(input, target);
expected = torch::tensor(0.3326, torch::kFloat);
s = output.sum();
s.backward();
ASSERT_TRUE(output.allclose(expected, 1e-04));
ASSERT_EQ(input.sizes(), input.grad().sizes());
// label smoothing with with target probabilities
loss = CrossEntropyLoss(CrossEntropyLossOptions().label_smoothing(0.2).reduction(torch::kMean));
input = torch::tensor({{3., 1.}, {1., 2.}}, torch::dtype(torch::kFloat).requires_grad(true));
target = torch::tensor({{0.8, 0.2}, {0.1, 0.9}}, torch::kFloat);
output = loss->forward(input, target);
expected = torch::tensor(0.5701, torch::kFloat);
s = output.sum();
s.backward();
ASSERT_TRUE(output.allclose(expected, 1e-04));
ASSERT_EQ(input.sizes(), input.grad().sizes());
}
TEST_F(ModulesTest, CosineSimilarity) {