We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
Updated ROCm PyTorch unit tests status (markdown)
Update url
Update HUD link
Update unit test status on ROCm post 2.7 enablement.
whitespace fixes
Update tests for ROCm 2.1
Update test counts based on latest CI run
Update test_sparse breakup
Update test_sparse numbers
After merging #248 by Chaitanya to enable more cuda tests.
Update all test set results after merging #247 from Chaitanya
test_sparse enablement got merged. Unit tests now overall at 89.6% (it's a lower bound as we may be too pessimistically counting skips)
Update some outdated fp16 info
After LLVM7 intro as base compiler, Chaitanya removed skips for min/max in test_cuda that now work.
update totals as well
test_utils finally got enabled
test_legacy_nn is no more - update test passing while here.
two more unit tests could be enabled by Chaitanya.
Update with test_sparse status
with latest update from Chaitanya
Fix compution error.
Enabling of tests in test_cuda and test_torch after compiler and HIP fixes.
Enabling CPU LAPACK related tests in test_distributions.
Update with the skips that seem unrelated to ROCm (skipped on CUDA, CPU only, flaky, ...)
Also add to total row
fix whitespace
Add a column for skipped on GPUs - we cannot reasonably assume to pass these tests and hence define pass rate accordingly. No data change yet.