dpctl

dpctl copied to clipboard

out of bound inputs for `uint8` and `uint16` data types

1

The following cases behave differently in `dpctl` and `NumPy`. ``` import dpctl,dpctl.tensor as dpt, numpy # uint8, with floating point input >>> a=numpy.array(60025, dtype=numpy.float32) >>> a.astype(numpy.uint8) array(121, dtype=uint8) >>> a=dpt.asarray(60025,...

vtavana

dpt.pow() with dtype=`c16` and scalar on gpu/cpu returns different result

2

The below example works differently depending on the device: ``` import dpctl.tensor as dpt a = dpt.asarray([0], dtype='c16', device='gpu') dpt.pow(a,1) # usm_ndarray([0.+0.j]) a = dpt.asarray([0], dtype='c16', device='cpu') dpt.pow(a,1) # usm_ndarray([nan+nanj])...

vlad-perevezentsev

__dlpack_device__() returned numbers

7

Hi, I have question about dlpack results. I created dpnp array, then checked \_\_dlpack_device\_\_() and got DLDeviceType=14(kDLOneAPI) and device_id =3. Could you help me understand what this 3 means? Because...

wozna

Array printing fails for `ndim` > `NPY_MAX_DIMS` with misleading `ValueError`

1

When printing an array `x` with `x.ndim > NPY_MAX_DIMS ` the following exception is raised: ``` In [1]: import dpctl.tensor as dpt, numpy as np In [2]: dpt.ones((1,) * 32)...

ndgrigorian

sycl::vec overloads for elementwise functions

10

This PR adds overloads for ``sycl::vec`` input to the implementations of ``dpctl.tensor.abs``, ``dpctl.tensor.cos``, ``dpctl.tensor.expm1``, ``dpctl.tensor.log``, ``dpctl.tensor.log1p``, and ``dpctl.tensor.sqrt``. - [X] Have you provided a meaningful PR description? - [X] Have...

ndgrigorian

dpctl should provide cmake config files and install them

2

The `FindDpctl.cmake` is not installed as part of the conda package nor do we currently provide a `DpctlConfig.cmake` file. The result is that dpctl cannot be found by external cmake...

diptorupd

ONEAPI_DEVICE_SELECTOR not working under sub-process

4

My environment: ``` bash export ONEAPI_DEVICE_SELECTOR="level_zero:gpu" ``` ``` bash $ sycl Warning: ONEAPI_DEVICE_SELECTOR environment variable is set to level_zero:gpu. To see the correct device id, please unset ONEAPI_DEVICE_SELECTOR. [ext_oneapi_level_zero:gpu:0] Intel(R)...

harborn

Custom implementation of isfinite

11

This PR should address issues test failures from gh-1378 It works around very annoying SYCLOS bug introduced with pull down of clang 18.0 It adds a custom implementation of `math_utils::isfinite`...

oleksandr-pavlyk

dpctl.tensor.put with mode="wrap" and negative indexes

2

The below example works differently with numpy: ```python a = dpt.zeros(7) ind = dpt.asarray([-8]) dpt.put(a, ind, 2, mode="wrap") a # Out: usm_ndarray([2., 0., 0., 0., 0., 0., 0.], dtype=float32) b...

antonwolfy

Implement `dpctl.tensor.trace` per array API [spec](https://data-apis.org/array-api/2021.12/extensions/generated/array_api.linalg.trace.html). This could be an all top-level function: 1. Form a view into diagonals 2. Call `dpctl.tensor.sum` on the last axis of the view Possible...

oleksandr-pavlyk