array-api-tests
array-api-tests copied to clipboard
data-apis
Test suite for the PyData Array APIs standard
Running individual tests of the test suite may lead to runs where no warnings are produced, and that triggers an exception. This change handles this corner case.
The tests run fast enough now that there's no need to lower this to 20. See the discussion at #246.
We need a tracking issue for 2023.12 support in the test suite. Here is the list of changes https://data-apis.org/array-api/latest/changelog.html#v2023-12 **New Functions** - [x] `__array_namespace_info__` (i.e., the [inspection API](https://data-apis.org/array-api/latest/API_specification/inspection.html)) - [x]...
Currently the suite fails when signatures use different arg names then whats in the spec for pos-only args, even though it shouldn't really matter. For example, testing a function with...
Per https://github.com/data-apis/array-api/pull/766#issuecomment-2018281825, not to merge yet as that PR isn't merge yet, and this is more of a POC to rework into the primary indexing tests possibly.
Running array-api-tests suite on `dpctl.tensor` (https://github.com/IntelPython/dpctl): ```bash ONEAPI_DEVICE_SELECTOR=*:cpu ARRAY_API_TESTS_MODULE=dpctl.tensor python -m pytest array_api_tests/test_creation_functions.py ``` two test new failures consistently occur (after updating from f82c7bc8627cc2c3a44fa3e425f53a253a609aa8 to recent tip of main, i.e....
The asarray test seems to only test the copy flag for array inputs. But it should also work for Python built-in (scalar or sequence) and buffer protocol inputs. copy=False should...
As far as I can tell `__array_namespace__` is not being tested at all. https://github.com/search?q=repo%3Adata-apis%2Farray-api-tests%20__array_namespace__&type=code
- Firstly we really need to - [x] Replace `numpy.array_api` with [`array-api-strict`](https://github.com/data-apis/array-api-strict) - `numpy.array_api` is going to be removed and supplanted by this third-party library - As bought up in...
This test fails with pytorch's sum() (note you need to use ``` E AssertionError: out=5.0, but should be roughly 7.0 [sum()] E Falsifying example: test_sum( E x=tensor([[ 0.0000e+00, 3.3554e+07, -3.3554e+07],...