New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[PyTorch] Add Vulkan support and tests for at::upsample_bilinear2d #98022
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/98022
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit 4f73c5e: This comment was automatically generated by Dr. CI and updates every 15 minutes. |
This pull request was exported from Phabricator. Differential Revision: D43142564 |
…ytorch#98022) Summary: Pull Request resolved: pytorch#98022 Bilinear upsampling is a [4D tensor upsampling operation](https://pytorch.org/docs/stable/generated/torch.nn.Upsample.html), this adds support for the operation on the Vulkan GPU backend. Test Plan: 1. `buck run --target-platforms ovr_config//platform/macos:arm64-fbsource //xplat/caffe2:pt_vulkan_api_test_binAppleMac\#macosx-arm64 -c pt.vulkan_full_precision=1` on Apple M1 MacBook 2. Confirm all tests pass with no regression, and the added tests `*upsample_bilinear2d*` pass 2a. All tests P669847383 2b. `upsample_bilinear2d` tests P669866631 3. Overview: ``` ... [ RUN ] VulkanAPITest.upsample_bilinear2d_align_false_small [ OK ] VulkanAPITest.upsample_bilinear2d_align_false_small (1 ms) [ RUN ] VulkanAPITest.upsample_bilinear2d_align_false_large [ OK ] VulkanAPITest.upsample_bilinear2d_align_false_large (2 ms) [ RUN ] VulkanAPITest.upsample_bilinear2d_align_true_small [ OK ] VulkanAPITest.upsample_bilinear2d_align_true_small (2 ms) [ RUN ] VulkanAPITest.upsample_bilinear2d_align_true_large [ OK ] VulkanAPITest.upsample_bilinear2d_align_true_large (1 ms) ... [==========] 209 tests from 1 test suite ran. (6317 ms total) [ PASSED ] 201 tests. [ SKIPPED ] 1 test, listed below: [ SKIPPED ] VulkanAPITest.querypool_flushed_shader_log [ FAILED ] 7 tests, listed below: [ FAILED ] VulkanAPITest.cat_dim1_singledepth_success [ FAILED ] VulkanAPITest.gru_success [ FAILED ] VulkanAPITest.gru_mclareninputs_success [ FAILED ] VulkanAPITest.gru_prepack_success [ FAILED ] VulkanAPITest.lstm_success [ FAILED ] VulkanAPITest.lstm_mclareninputs_success [ FAILED ] VulkanAPITest.lstm_prepack_success ``` Reviewed By: SS-JIA Differential Revision: D43142564 fbshipit-source-id: 39931862c2700e69562565042e2f9e92a262f276
This pull request was exported from Phabricator. Differential Revision: D43142564 |
32b23df
to
4f73c5e
Compare
@pytorchbot merge (Initiating merge automatically since Phabricator Diff has merged) |
@pytorchbot merge (Initiating merge automatically since Phabricator Diff has merged) |
Merge startedYour change will be merged once all checks pass (ETA 0-4 Hours). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
Summary: Bilinear upsampling is a 4D tensor upsampling operation, this adds support for the operation on the Vulkan GPU backend.
Test Plan:
buck run --target-platforms ovr_config//platform/macos:arm64-fbsource //xplat/caffe2:pt_vulkan_api_test_binAppleMac\#macosx-arm64 -c pt.vulkan_full_precision=1
on Apple M1 MacBook*upsample_bilinear2d*
pass2a. All tests P669847383
2b.
upsample_bilinear2d
tests P669866631Differential Revision: D43142564