-
Notifications
You must be signed in to change notification settings - Fork 17
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
Add complex f32 f64 (release-staging/rocm-rel-6.1) #175
Add complex f32 f64 (release-staging/rocm-rel-6.1) #175
Commits on Nov 30, 2023
-
Bump rocm-docs-core from 0.28.0 to 0.30.0 in /docs/.sphinx
Bumps [rocm-docs-core](https://github.com/RadeonOpenCompute/rocm-docs-core) from 0.28.0 to 0.30.0. - [Release notes](https://github.com/RadeonOpenCompute/rocm-docs-core/releases) - [Changelog](https://github.com/RadeonOpenCompute/rocm-docs-core/blob/develop/CHANGELOG.md) - [Commits](ROCm/rocm-docs-core@v0.28.0...v0.30.0) --- updated-dependencies: - dependency-name: rocm-docs-core dependency-type: direct:production update-type: version-update:semver-minor ... Signed-off-by: dependabot[bot] <support@github.com>
Configuration menu - View commit details
-
Copy full SHA for 1f69024 - Browse repository at this point
Copy the full SHA 1f69024View commit details
Commits on Dec 4, 2023
-
Merge pull request ROCm#161 from ROCmSoftwarePlatform/dependabot/pip/…
…docs/dot-sphinx/rocm-docs-core-0.30.0 Bump rocm-docs-core from 0.28.0 to 0.30.0 in /docs/.sphinx
Configuration menu - View commit details
-
Copy full SHA for d71dcad - Browse repository at this point
Copy the full SHA d71dcadView commit details
Commits on Dec 6, 2023
-
Rename `permutation` to `simple_permutation` to comply with naming conventions.
Configuration menu - View commit details
-
Copy full SHA for fc450d9 - Browse repository at this point
Copy the full SHA fc450d9View commit details -
Add unittest to for the permute CPU implementation
- compare the permute result with reference - test col/row major - test float and _Float16
Configuration menu - View commit details
-
Copy full SHA for 9c4c762 - Browse repository at this point
Copy the full SHA 9c4c762View commit details -
Bump gitpython from 3.1.35 to 3.1.37 in /docs/.sphinx
Bumps [gitpython](https://github.com/gitpython-developers/GitPython) from 3.1.35 to 3.1.37. - [Release notes](https://github.com/gitpython-developers/GitPython/releases) - [Changelog](https://github.com/gitpython-developers/GitPython/blob/main/CHANGES) - [Commits](gitpython-developers/GitPython@3.1.35...3.1.37) --- updated-dependencies: - dependency-name: gitpython dependency-type: indirect ... Signed-off-by: dependabot[bot] <support@github.com>
Configuration menu - View commit details
-
Copy full SHA for bba3217 - Browse repository at this point
Copy the full SHA bba3217View commit details -
Bump urllib3 from 1.26.15 to 1.26.18 in /docs/.sphinx
Bumps [urllib3](https://github.com/urllib3/urllib3) from 1.26.15 to 1.26.18. - [Release notes](https://github.com/urllib3/urllib3/releases) - [Changelog](https://github.com/urllib3/urllib3/blob/main/CHANGES.rst) - [Commits](urllib3/urllib3@1.26.15...1.26.18) --- updated-dependencies: - dependency-name: urllib3 dependency-type: indirect ... Signed-off-by: dependabot[bot] <support@github.com>
Configuration menu - View commit details
-
Copy full SHA for 3f7a904 - Browse repository at this point
Copy the full SHA 3f7a904View commit details
Commits on Dec 7, 2023
-
Bump rocm-docs-core from 0.30.0 to 0.30.1 in /docs/.sphinx
Bumps [rocm-docs-core](https://github.com/RadeonOpenCompute/rocm-docs-core) from 0.30.0 to 0.30.1. - [Release notes](https://github.com/RadeonOpenCompute/rocm-docs-core/releases) - [Changelog](https://github.com/RadeonOpenCompute/rocm-docs-core/blob/develop/CHANGELOG.md) - [Commits](ROCm/rocm-docs-core@v0.30.0...v0.30.1) --- updated-dependencies: - dependency-name: rocm-docs-core dependency-type: direct:production update-type: version-update:semver-patch ... Signed-off-by: dependabot[bot] <support@github.com>
Configuration menu - View commit details
-
Copy full SHA for 23b46d6 - Browse repository at this point
Copy the full SHA 23b46d6View commit details -
Merge pull request ROCm#156 from CongMa13/permute_cpu_unit_test
Add unittest to for the permute CPU implementation
Configuration menu - View commit details
-
Copy full SHA for 0be15e6 - Browse repository at this point
Copy the full SHA 0be15e6View commit details -
Fix spelling in documentation (ROCm#155)
* Fix spelling in documentation * Use code directive to escape code keywords * Revert "Use code directive to escape code keywords" This reverts commit 7be7e34. * Disable spellcheck on API Reference Guide * Fix spelling in API Reference Guide hiptensor > hipTensor
Configuration menu - View commit details
-
Copy full SHA for 852992e - Browse repository at this point
Copy the full SHA 852992eView commit details
Commits on Dec 8, 2023
-
Add support to f16 and bf16 to contraction
- Support _Float16 - Support hip_bfloat16 - Add unit test of _Float16 and hip_bfloat16 - Add sample of _Float16 and hip_bfloat16
Configuration menu - View commit details
-
Copy full SHA for c5fbcec - Browse repository at this point
Copy the full SHA c5fbcecView commit details -
Add support to f32_f16, f32_bf16, f64_f32 to contraction
- Support ABCD data type f32 and compute type f16, bf16 - Support ABCD data type f64 and compute type f32 - Fixed bug: alpha, beta were passed in as wrong data type in unit test of contraction - Create sample template of contraction
Configuration menu - View commit details
-
Copy full SHA for 185a2ab - Browse repository at this point
Copy the full SHA 185a2abView commit details -
Add placeholder for solution unique_id
Solution unique_ids of Actor Critic are have not been ready yet, but we put some placeholders in the new Actor Critic to make the unit tests be able to pass.
Configuration menu - View commit details
-
Copy full SHA for ab8d557 - Browse repository at this point
Copy the full SHA ab8d557View commit details -
Update contraction device instances
Update contraction device instances since CK has updated them.
Configuration menu - View commit details
-
Copy full SHA for df27e32 - Browse repository at this point
Copy the full SHA df27e32View commit details -
1. Initiate the data with 0.01, 0.02, ... by default 2. Print C
Configuration menu - View commit details
-
Copy full SHA for f85df83 - Browse repository at this point
Copy the full SHA f85df83View commit details -
Set CK contraction instance only run once
When logger level is set to HIPTENSOR_LOG_LEVEL_PERF_TRACE, we make CK instances measure the running time. The problem is that CK internally will run the contraction 10 times by default. This leads to an issues: 1. It returns wrong result for C = alpha A x B + beta C Set StreamConfig.nrepeat_ = 1, the contraction will be run once
Configuration menu - View commit details
-
Copy full SHA for 5c45a8c - Browse repository at this point
Copy the full SHA 5c45a8cView commit details -
1. ck::bhalf_t cannot cast to float or double by static_cast. Use ck::type_convert() to fix it. 2. epsilon() is not good value to measure the relative difference of data. It is too small for double (eps < 10e-13).
Configuration menu - View commit details
-
Copy full SHA for f631818 - Browse repository at this point
Copy the full SHA f631818View commit details -
Configuration menu - View commit details
-
Copy full SHA for e5cefe7 - Browse repository at this point
Copy the full SHA e5cefe7View commit details -
Rename contraction sameple files
The pattern of contraction sameple file is - bilinear: simple_bilinear_contraction_<A>_<B>_<C>_<D>_compute_<compute>.cpp - scale : simple_scale_contraction_<A>_<B>_<C>_compute_<compute>.cpp
Configuration menu - View commit details
-
Copy full SHA for 4345a1c - Browse repository at this point
Copy the full SHA 4345a1cView commit details -
Improve CPU reference accurary
The relative difference between contraction result and CPU reference is less than 0.1% after the improvement.
Configuration menu - View commit details
-
Copy full SHA for 43f33ee - Browse repository at this point
Copy the full SHA 43f33eeView commit details -
Configuration menu - View commit details
-
Copy full SHA for fec9065 - Browse repository at this point
Copy the full SHA fec9065View commit details -
1. Revert the default threshold of relative difference to (100 * std::numeric_limits<T>::epsilon()) 2. Update CPU reference to make the difference between CPU reference and output of contraction instance is less than (100 * std::numeric_limits<T>::epsilon()).
Configuration menu - View commit details
-
Copy full SHA for b21fe0b - Browse repository at this point
Copy the full SHA b21fe0bView commit details
Commits on Dec 11, 2023
-
Merge pull request ROCm#158 from CongMa13/contraction_f16_bf16
Contraction f16, bf16, f32_f16, f32_bf16, f64_f32
Configuration menu - View commit details
-
Copy full SHA for 8c11d59 - Browse repository at this point
Copy the full SHA 8c11d59View commit details -
Configuration menu - View commit details
-
Copy full SHA for 76de7d0 - Browse repository at this point
Copy the full SHA 76de7d0View commit details -
Merge pull request ROCm#166 from ROCm/dependabot/pip/docs/dot-sphinx/…
…rocm-docs-core-0.30.1 Bump rocm-docs-core from 0.30.0 to 0.30.1 in /docs/.sphinx
Configuration menu - View commit details
-
Copy full SHA for 71ae65d - Browse repository at this point
Copy the full SHA 71ae65dView commit details -
Merge pull request ROCm#164 from ROCm/dependabot/pip/docs/dot-sphinx/…
…urllib3-1.26.18 Bump urllib3 from 1.26.15 to 1.26.18 in /docs/.sphinx
Configuration menu - View commit details
-
Copy full SHA for 1dfaea7 - Browse repository at this point
Copy the full SHA 1dfaea7View commit details -
Merge pull request ROCm#163 from ROCm/dependabot/pip/docs/dot-sphinx/…
…gitpython-3.1.37 Bump gitpython from 3.1.35 to 3.1.37 in /docs/.sphinx
Configuration menu - View commit details
-
Copy full SHA for b88175f - Browse repository at this point
Copy the full SHA b88175fView commit details
Commits on Dec 12, 2023
-
Merge pull request ROCm#169 from CongMa13/remove_xfloat32
Remove xfloat32 which is not used in hiptensor
Configuration menu - View commit details
-
Copy full SHA for 53d687b - Browse repository at this point
Copy the full SHA 53d687bView commit details -
Configuration menu - View commit details
-
Copy full SHA for 28fe756 - Browse repository at this point
Copy the full SHA 28fe756View commit details -
Configuration menu - View commit details
-
Copy full SHA for c18335a - Browse repository at this point
Copy the full SHA c18335aView commit details -
Bump cryptography from 41.0.4 to 41.0.6 in /docs/.sphinx (ROCm#162)
Bumps [cryptography](https://github.com/pyca/cryptography) from 41.0.4 to 41.0.6. - [Changelog](https://github.com/pyca/cryptography/blob/main/CHANGELOG.rst) - [Commits](pyca/cryptography@41.0.4...41.0.6) --- updated-dependencies: - dependency-name: cryptography dependency-type: indirect ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
Configuration menu - View commit details
-
Copy full SHA for 7ac3fb9 - Browse repository at this point
Copy the full SHA 7ac3fb9View commit details
Commits on Dec 13, 2023
-
Configuration menu - View commit details
-
Copy full SHA for b03e4f3 - Browse repository at this point
Copy the full SHA b03e4f3View commit details
Commits on Dec 14, 2023
-
Configuration menu - View commit details
-
Copy full SHA for ded69b9 - Browse repository at this point
Copy the full SHA ded69b9View commit details -
Merge pull request ROCm#160 from StreamHPC/docs_fix
Fix docs build error and warnings
Configuration menu - View commit details
-
Copy full SHA for bcd81bd - Browse repository at this point
Copy the full SHA bcd81bdView commit details
Commits on Dec 20, 2023
-
Bump rocm-docs-core from 0.30.1 to 0.30.2 in /docs/.sphinx (ROCm#171)
Bumps [rocm-docs-core](https://github.com/RadeonOpenCompute/rocm-docs-core) from 0.30.1 to 0.30.2. - [Release notes](https://github.com/RadeonOpenCompute/rocm-docs-core/releases) - [Changelog](https://github.com/RadeonOpenCompute/rocm-docs-core/blob/develop/CHANGELOG.md) - [Commits](ROCm/rocm-docs-core@v0.30.1...v0.30.2) --- updated-dependencies: - dependency-name: rocm-docs-core dependency-type: direct:production update-type: version-update:semver-patch ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
Configuration menu - View commit details
-
Copy full SHA for 0b34d47 - Browse repository at this point
Copy the full SHA 0b34d47View commit details -
Bump rocm-docs-core from 0.30.2 to 0.30.3 in /docs/.sphinx
Bumps [rocm-docs-core](https://github.com/RadeonOpenCompute/rocm-docs-core) from 0.30.2 to 0.30.3. - [Release notes](https://github.com/RadeonOpenCompute/rocm-docs-core/releases) - [Changelog](https://github.com/RadeonOpenCompute/rocm-docs-core/blob/develop/CHANGELOG.md) - [Commits](ROCm/rocm-docs-core@v0.30.2...v0.30.3) --- updated-dependencies: - dependency-name: rocm-docs-core dependency-type: direct:production update-type: version-update:semver-patch ... Signed-off-by: dependabot[bot] <support@github.com>
Configuration menu - View commit details
-
Copy full SHA for 95af3c1 - Browse repository at this point
Copy the full SHA 95af3c1View commit details -
Merge pull request ROCm#173 from ROCm/dependabot/pip/docs/dot-sphinx/…
…rocm-docs-core-0.30.3 Bump rocm-docs-core from 0.30.2 to 0.30.3 in /docs/.sphinx
Configuration menu - View commit details
-
Copy full SHA for f5803f9 - Browse repository at this point
Copy the full SHA f5803f9View commit details
Commits on Jan 2, 2024
-
Configuration menu - View commit details
-
Copy full SHA for e41eda6 - Browse repository at this point
Copy the full SHA e41eda6View commit details
Commits on Jan 4, 2024
-
- Add API changes - Add test files - Add samples - Add cpu contraction for complex types - Add complex instances
Configuration menu - View commit details
-
Copy full SHA for 7496442 - Browse repository at this point
Copy the full SHA 7496442View commit details -
- Add compute types - Add instance factory for scale
Configuration menu - View commit details
-
Copy full SHA for be8c0b5 - Browse repository at this point
Copy the full SHA be8c0b5View commit details -
- Modify complex scale impl - Modify pack func
Configuration menu - View commit details
-
Copy full SHA for f6e39c1 - Browse repository at this point
Copy the full SHA f6e39c1View commit details -
- Remove intermediate buffers
Configuration menu - View commit details
-
Copy full SHA for 075cdfa - Browse repository at this point
Copy the full SHA 075cdfaView commit details -
Add scalar and complex type instances
- Add cpu instances - Add scale complex initial imp - Remove compute f32 instances for complex f32 - Modify multiply utility - Modified bilinear to take complex compute
Configuration menu - View commit details
-
Copy full SHA for 1bf4a55 - Browse repository at this point
Copy the full SHA 1bf4a55View commit details -
Configuration menu - View commit details
-
Copy full SHA for b50a159 - Browse repository at this point
Copy the full SHA b50a159View commit details -
Configuration menu - View commit details
-
Copy full SHA for b85aa93 - Browse repository at this point
Copy the full SHA b85aa93View commit details -
Configuration menu - View commit details
-
Copy full SHA for 1b2031e - Browse repository at this point
Copy the full SHA 1b2031eView commit details -
Configuration menu - View commit details
-
Copy full SHA for 41c44b2 - Browse repository at this point
Copy the full SHA 41c44b2View commit details -
Add unit test of contraction with complex compute type
- Fix issue in yaml_test - Fix a bug in hiptensorInitContractionDescriptor
Configuration menu - View commit details
-
Copy full SHA for cf233d8 - Browse repository at this point
Copy the full SHA cf233d8View commit details -
Modify samples to use new compute type
- Fix bug in samples - Add unit tests for scale contraction
Configuration menu - View commit details
-
Copy full SHA for c01eda7 - Browse repository at this point
Copy the full SHA c01eda7View commit details -
Support Complex alpha and beta in samples
- Add complex_scale_contraction_test.cpp - Fixed bug in device_contraction_bilinear_complex.hpp
Configuration menu - View commit details
-
Copy full SHA for 23c033f - Browse repository at this point
Copy the full SHA 23c033fView commit details -
Configuration menu - View commit details
-
Copy full SHA for 48abe4a - Browse repository at this point
Copy the full SHA 48abe4aView commit details -
Set unit test difference threshold to epsilon of compute type
- New single kernel selection. To be improved. - Used instance selected by brute force to compute tensor with 1 as the most right stride - Fixed bug that used data type id as compute data type id
Configuration menu - View commit details
-
Copy full SHA for 348e281 - Browse repository at this point
Copy the full SHA 348e281View commit details