OSSForks/pytorch
1
0
Fork 0
mirror of https://github.com/zebrajr/pytorch.git synced 2025-12-06 12:20:52 +01:00
Code Issues Actions 38 Packages Projects Releases Wiki Activity

Move slow-grad checks to CUDA-11.6 (#84313)

Browse Source 
Mitigates #84192 by skipping two tests

Please note: We tried to increase the tolerance for test_fn_gradgrad_linalg_det_singular_cuda_float64 but this did not help.
Ref:
Increase `test_fn_gradgrad_linalg_det_singular_cuda_float64` error tolerance to  1e-4 as suggested in https://github.com/pytorch/pytorch/issues/84192#issuecomment-1230644574

Pull Request resolved: https://github.com/pytorch/pytorch/pull/84313
Approved by: https://github.com/malfet, https://github.com/huydhn, https://github.com/Lezcano
This commit is contained in:
Andrey Talman 2022-09-01 20:24:06 +00:00 committed by PyTorch MergeBot
parent 673b35c847
commit 5e5c610a58
2 changed files with 21 additions and 16 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

25
.github/workflows/periodic.yml vendored
View File

@ -13,20 +13,20 @@ concurrency:
cancel-in-progress: true
jobs:
linux-xenial-cuda10_2-py3-gcc7-slow-gradcheck-build:
name: linux-xenial-cuda10.2-py3-gcc7-slow-gradcheck
linux-bionic-cuda11_6-py3-gcc7-slow-gradcheck-build:
name: linux-bionic-cuda11.6-py3-gcc7-slow-gradcheck
uses: ./.github/workflows/_linux-build.yml
with:
build-environment: linux-xenial-cuda10.2-py3-gcc7-slow-gradcheck
docker-image-name: pytorch-linux-xenial-cuda10.2-cudnn7-py3-gcc7
build-environment: linux-bionic-cuda11.6-py3-gcc7-slow-gradcheck
docker-image-name: pytorch-linux-bionic-cuda11.6-cudnn8-py3-gcc7
linux-xenial-cuda10_2-py3-gcc7-slow-gradcheck-test:
name: linux-xenial-cuda10.2-py3-gcc7-slow-gradcheck
linux-bionic-cuda11_6-py3-gcc7-slow-gradcheck-test:
name: linux-bionic-cuda11.6-py3-gcc7-slow-gradcheck
uses: ./.github/workflows/_linux-test.yml
needs: linux-xenial-cuda10_2-py3-gcc7-slow-gradcheck-build
needs: linux-bionic-cuda11_6-py3-gcc7-slow-gradcheck-build
with:
build-environment: linux-xenial-cuda10.2-py3-gcc7-slow-gradcheck
docker-image: ${{ needs.linux-xenial-cuda10_2-py3-gcc7-slow-gradcheck-build.outputs.docker-image }}
build-environment: linux-bionic-cuda11.6-py3-gcc7-slow-gradcheck
docker-image: ${{ needs.linux-bionic-cuda11_6-py3-gcc7-slow-gradcheck-build.outputs.docker-image }}
test-matrix: |
{ include: [
{ config: "default", shard: 1, num_shards: 2, runner: "linux.4xlarge.nvidia.gpu" },
@ -78,13 +78,6 @@ jobs:
AWS_OSSCI_METRICS_V2_ACCESS_KEY_ID: ${{ secrets.AWS_OSSCI_METRICS_V2_ACCESS_KEY_ID }}
AWS_OSSCI_METRICS_V2_SECRET_ACCESS_KEY: ${{ secrets.AWS_OSSCI_METRICS_V2_SECRET_ACCESS_KEY }}
linux-bionic-cuda10_2-py3_9-gcc7-build:
name: linux-bionic-cuda10.2-py3.9-gcc7
uses: ./.github/workflows/_linux-build.yml
with:
build-environment: linux-bionic-cuda10.2-py3.9-gcc7
docker-image-name: pytorch-linux-bionic-cuda10.2-cudnn7-py3.9-gcc7
linux-bionic-cuda11_6-py3_9-gcc7-build:
name: linux-bionic-cuda11.6-py3.9-gcc7
uses: ./.github/workflows/_linux-build.yml

12
torch/testing/_internal/opinfo/definitions/linalg.py
View File

@ -1312,6 +1312,18 @@ op_db: List[OpInfo] = [
"test_fn_fwgrad_bwgrad",
dtypes=(torch.complex128,),
),
DecorateInfo(
unittest.skip("Skipped, see https://github.com//issues/84192"),
"TestGradients",
"test_fn_gradgrad",
device_type="cuda",
),
DecorateInfo(
unittest.skip("Skipped, see https://github.com//issues/84192"),
"TestGradients",
"test_fn_fwgrad_bwgrad",
device_type="cuda",
),
),
),
OpInfo(