saymrwulf/pytorch
1
0
Fork 0
mirror of https://github.com/saymrwulf/pytorch.git synced 2026-05-14 20:57:59 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

[SR] Fix quantized linear tests not managing outputs (#75776)

Browse source 
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/75776

The output was returned directly instead of a clone, so the output of the relevant op would not be managed.
ghstack-source-id: 154935103

Test Plan: CI

Reviewed By: navahgar

Differential Revision: D35633469

fbshipit-source-id: 7b08b7368e0349a12abf8802a4c625ffecdc5abb
(cherry picked from commit 24bed9ba4da39cff7f3b40f5e49dfded2552b373)
This commit is contained in:
Mike Iovine 2022-04-27 09:32:27 -07:00 committed by PyTorch MergeBot
parent 04b3313379
commit 3fa77fa51a
1 changed files with 2 additions and 2 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

4
benchmarks/static_runtime/test_static_runtime.cc
View file

@ -2133,7 +2133,7 @@ TEST(StaticRuntime, QuantizedLinearDynamicFp16) {
%packed_params = quantized::linear_prepack_fp16(%weights, %bias)
%output = quantized::linear_dynamic_fp16(%input, %packed_params)
%ret = aten::clone(%output, %bias)
return (%output)
return (%ret)
)IR";
at::Tensor weight = torch::randn({3, 2}, torch::kFloat);
at::Tensor input = torch::randn({3, 2}, torch::kFloat);
@ -2154,7 +2154,7 @@ TEST(StaticRuntime, QuantizedLinearReluDynamicFp16) {
%packed_params = quantized::linear_prepack_fp16(%weights, %bias)
%output = quantized::linear_relu_dynamic_fp16(%input, %packed_params)
%ret = aten::clone(%output, %bias)
return (%output)
return (%ret)
)IR";
at::Tensor weight = torch::randn({3, 2}, torch::kFloat);
at::Tensor input = torch::randn({3, 2}, torch::kFloat);