dpctl
dpctl copied to clipboard
Python SYCL bindings and SYCL-based Python Array API library
In this github action example, the conda environment can properly access the installed opencl driver, but pip cannot: ``` name: pip dpctl validator on: workflow_dispatch: env: PYTHON_VERSION: 3.9 DPCTL_VERSION: 0.17.0...
The below example reproduces an issue with passing integer scalar to the shape setter: ```python import numpy, dpctl, dpctl.tensor as dpt dpctl.__version__ # Out: '0.18.0dev0+158.g7450558d25' a = dpt.usm_ndarray((2, 3)) a.shape...
dpctl tensors implement most of the array_api standard except for the linalg package as shown by this deselection file: https://github.com/IntelPython/dpctl/blob/master/.github/workflows/array-api-skips.txt This causes issues in using dpctl tensors with scikit-learn, an...
This PR modifies cmake scripts throughout dpctl to enable building for AMD. This is done by either setting the `DPCTL_TARGET_AMD` environment variable to the intended build architecture, or using `-DDPCTL_TARGET_AMD`....
The in-place elementwise operator cannot operate on the largest dtype for the second value. ``` python >>> import dpctl.tensor as dpt >>> a = dpt.asarray(10, dtype="f4", device="cpu") >>> b =...
Currently SyclQueue is not picklable. The error message is the following: ``` File "", line 2, in dpctl._sycl_queue.SyclQueue.__reduce_cython__ TypeError: no default __reduce__ due to non-trivial __cinit__ ``` The queue is...
The below example: ```python import dpnp, dpctl, dpctl.tensor as dpt dpctl.__version__ # Out: '0.18.0dev0+158.g7450558d25' a = dpnp.ones(10) a.flags # Out: # C_CONTIGUOUS : True # F_CONTIGUOUS : True # WRITABLE...
System info: ``` OS Name: Microsoft Windows 11 Enterprise OS Version: 10.0.22631 N/A Build 22631 System Type: x64-based PC Processor(s): 1 Processor(s) Installed. [01]: Intel64 Family 6 Model 140 Stepping...
The input is ```python import dpctl.tensor as dpt val = dpt.zeros(2, dtype="u1") max = dpt.zeros(tuple(), dtype="i1") dpt.clip(val, min=None, max=max) ``` Raises `ValueError: function 'clip' does not support input types (uint8,...
Current implementations of elementwise functions in dpctl.tensor do not support `out` data type other than `bool` and raise `ValueError` if `out` is not `bool`. ``` import dpctl.tensor as dpt a...