IntelPython / dpctl

Python SYCL bindings and SYCL-based Python Array API library

https://intelpython.github.io/dpctl/

IntelPython/dpctl Issues

Incorrect result in dpctl.tensor.tensordot for `axes=0`
Closed 3 months ago
Removal of libsyclinterface DPCTLQueueMgr_* function related to device_context
Updated 3 months ago4
Any/all reductions without atomic aspect
Closed 4 months ago1
dpctl.tensor.arcsin with c16 dtype crashes on H100
Closed 4 months ago3
out of bound inputs for `uint8` and `uint16` data types
Updated 4 months ago1
Element-wise function performance has degraded
Closed 4 months ago1
tensor.asarray raises unexpected TypeError for out of bound inputs
Closed 5 months ago1
Implement `device` keyword into `astype`
Closed 5 months ago1
dpctl could not properly handle NaN edges in `clip` function
Closed 5 months ago1
Wrong element copy for 2D array
Closed 5 months ago
dpt.pow() with dtype=`c16` and scalar on gpu/cpu returns different result
Updated 5 months ago2
__dlpack_device__() returned numbers
Updated 5 months ago7
`test_projection_complex[c16]` fails with the latest version of the compiler [CPU]
Closed 5 months ago3
Unable to build dpctl due to compilation errors in cython generated files
Closed 5 months ago7
Implement `device` keyword into `can_cast`, `result_type`
Closed 6 months ago3
discrepancy of results for `dpctl.tensor.sin` and `dpt.tensor.expm1` for `f4` and `f8` data types on `gpu`
Closed 6 months ago1
dpctl doesn't allow mixing integer dtypes of indices arrays
Updated 6 months ago
Array printing fails for `ndim` > 32 with misleading `ValueError`
Updated 6 months ago
dpctl does not return a zero-length array when no reduction is going to be done
Closed 7 months ago
dpt.sum is not doing computation in requested `dtype` on Windows
Closed 7 months ago
Sensible performance degradation in `dpt.tensor.sum`
Closed 7 months ago
Unhelpful error message when calling `asnumpy` with non-usm_array type
Closed 7 months ago1
`dpctl.tensor.max` returns incorrect result when reducing over `axis=0`
Closed 7 months ago
dpctl.tensor.sum function raises an error: SyclAsynchronousError
Closed 8 months ago4
Implement GH action to purge dppy/label/dev of old artifacts
Closed 8 months ago1
repeat function return incorrect result for axis 0
Closed 8 months ago
`test_atan2.py` fails with the latest version of the compiler
Closed 9 months ago2
elementwise test files are not being copied into the layout
Closed 9 months ago
ONEAPI_DEVICE_SELECTOR not working under sub-process
Updated 9 months ago4
Investigate failing SYCLOS workflow
Closed 9 months ago3
crash in test_sycl_kernel_submit.py
Closed 9 months ago5
dpt.sum over axis for transposed input array
Closed 9 months ago1
`dpt.put` returns a TypeError when `vals` is usm_ndarray with deferent dtype than `x`
Updated 9 months ago
output array with a data type higher in hierarchy
Closed 9 months ago2
lowercase order value in `dpctl.tensor.copy()`
Closed 9 months ago
dpctl.tensor.put with mode="wrap" and negative indexes
Updated 10 months ago1
dpctl.tensor.put returns incorrect result
Updated 10 months ago1
Remove deprecated DPCTLDevice_GetMaxWorkItemSizes
Closed 10 months ago
Implement tensor.trace
Updated 10 months ago
Copy usm_ndarray fails with order='K'
Closed 10 months ago
dpctl_capi does not compile with C compiler
Closed 10 months ago
Missing C- and F-contagious flags
Closed 10 months ago
Array assignment with broadcasting is not working
Closed 10 months ago2
dpctl.tensor.nonzero returns an incorrect array dtype in dpctl=0.14.6dev1 [win]
Closed 10 months ago1
Incorrect behavior of `order` parameter for `order="K"`
Closed 10 months ago2
dpctl.tensor.logaddexp does not always return `nan` when one operand is not `nan`
Closed 10 months ago2
dpctl.tensor.concat() failed on Iris Xe
Closed 10 months ago
Incorrect behavior of `dpctl.tensor.all` and `dpctl.tensor.any` with `axis=0` parameter
Closed 10 months ago
full and full_like do not work for `fill_value=dpt.nan`
Closed 10 months ago1
dpt.nonzero returns array of int32 dtype
Closed 10 months ago