Diptorup Deb
Diptorup Deb
- [X] Have you provided a meaningful PR description? sklearn-numba-dpex provides an implementation of random number generator functions using numba-dpex. The feature is analogous to similar functionality provided by numba.cuda....
I encountered a segfault when running the dpjit reduction test case on ARC with numba-dpex 0.23.0rc1. The same tests fail with 0.21.1, but do not segfault. ```bash $ ONEAPI_DEVICE_SELECTOR=*:gpu NUMBA_CAPTURED_ERRORS=new_style...
A meta issue to aggregate various numba-dpex documentation related issues in one place: - [x] #1085 - [x] Ensure no all code snippets in the documentation work and do not...
An implementation for the `static_getitem` function for `DpnpNdArray` is missing and that results in the lowering falling back to the generic `lower_getitem` that in turn resolves to the `getitem_arraynd_intp` call...
Numba-dpex only has end-to-end tests for kernels. There are no tests to validate the LLVM IR that is generated for a device with corresponding IR generated by dpcpp. There needs...
The `dpnp.asarray` constructor should be supported inside a `dpjit` decorated function.
Numba has a decorator [vectorize](https://numba.readthedocs.io/en/stable/user/vectorize.html#the-vectorize-decorator) that allows creation of a NumPy `ufunc`. The feature was previously partially supported in numba-dpex, but broke around the 0.20 release (refer: #807). The existing...
Right now numba-dpex serializes the LLVM bitcode to disk and then passes it to `llvm-spirv`. The serialization can increase dpex compile time. Numba-dpex should directly pass the LLVM bitcode to...
SYCL 2020 rev 8 provides a cross-platform [`vec` class template](https://registry.khronos.org/SYCL/specs/sycl-2020/html/sycl-2020.html#sec:vector.type) that works efficiently on SYCL devices. The issue is to open a discussion on how an analogous feature can be...
numba-dpex kernel API does not support the [oneAPI ESIMD extension](https://github.com/intel/llvm/tree/sycl/sycl/doc/extensions/supported/sycl_ext_intel_esimd#readme). The issue is to start a discussion if adding support for the extension is viable and useful for our kernel...