Franck Charras
Franck Charras
In https://benchopt.github.io/benchmark_workflow/write_benchmark.html#solver-from-source The example given does not install from source ? it's a excerpt with a regular install of scikit-learn from with conda. The same is given in the python...
Fixes https://github.com/joblib/joblib/issues/1545 Still todo: - [x] add non regression test from min reproducer in https://github.com/joblib/joblib/issues/1545 - [x] ~~discuss where exactly we should enable / disable this feature ? to what...
`math.ceil` and `math.floor` [are supported within a numba kernel](https://intelpython.github.io/numba-dpex/latest/user_guides/kernel_programming_guide/supported-python-features.html#standard-library-modules) and when using it within a `CPython` interpreter it returns an `int`. But within a `dpex.kernel` it returns a `float64`, which...
The documentation currently states that [`numba_dpex` does not yet support SYCL private and constant memory](https://intelpython.github.io/numba-dpex/latest/user_guides/kernel_programming_guide/memory-management.html#private-and-constant-memory), but it does in fact support private memory (`dpex.private.array`).
The following does not work: ```python import numba_dpex as dpex from numba import float32 import dpctl import dpnp import numpy as np @dpex.kernel def kernel(array_i): i = dpex.get_global_id(0) array[i] =...
The following snippet highlights differences regarding loop unrolling between `numba` and `numba_dpex`. Regular `numba` [will try to unroll loops](https://numba.pydata.org/numba-doc/latest/user/faq.html#why-my-loop-is-not-vectorized) but the same behavior is not seen with `numba_dpex`, as it...
https://github.com/IntelPython/numba-dpex/issues/159 was closed as completed after analogue to `np.random` functions were added to `dpnp` but it does not addresses some use cases that are met with `numba.cuda.random`: - generating `float32`...
There are several reasons I could think a reference matmul kernel implemented with `numba_dpex.kernel` is interesting: - avoid the need to use `dpnp.matmul`, which is one less dependency, and let...
The closer that we can get in `dpctl` seems to be `max_compute_units` but that does not seem to do the job, online documentation seems to suggest that the number of...
dpctl can detect cuda devices: ``` In [2]: import dpctl In [3]: dpctl.get_devices() Out[3]: [] ``` but `dpctl.program` can't create kernels, [only `level_zero` and `opencl` are supported](https://github.com/IntelPython/dpctl/blob/master/libsyclinterface/source/dpctl_sycl_kernel_bundle_interface.cpp#L586): Are there plans...