Mads R. B. Kristensen
Mads R. B. Kristensen
## Description This PR introduces factory functions to create `Buffer` instances, which makes it possible to change the returned buffer type based on a configuration option in a follow-up PR....
As far as I can see, we do not install the type stub files such as `column.pyi`. At least, building cudf and libcudf don't seem to install them? ``` ./build.sh...
This is part of the effort to implement seamlessly spilling in cuDF and is just for testing for now. The idea is to have a new column accessor, `SpillableColumnAccessor`, that...
The idea is to implement a Python interface with a minimal overhead. ### Preliminary performance Running [this small benchmark](https://github.com/madsbk/bohrium/blob/bohr/bridge/bohr/small_bench.py) produces on my laptop: ``` Numpy: 0.06131768226623535 Bohr: 0.4523584842681885 Bohrium: 8.404689311981201...
The current pip package does not include the visualize module. We should create a new pip package `bohrium-visualize` that adds visualize support and depend on glut.
CUDA and OpenCL has vector types such as `cl_float4` or `cl_int4`, which might improve performance when used.
At the moment, the greedy fuser re-calculates the fusibility between all nodes in each iteration. We should maintain the fusibility between iteration by only re-calculating the adjacent nodes.
This PR implements support of alternative array types (other than NumPy arrays) by introducing a new argument, `meta_array`, that specifies the type/class of the underlying array. The `meta_array` argument can...
This is based on #10746 and uses the newly refactored [`Buffer`](https://github.com/rapidsai/cudf/pull/11447). We introduce a new deivce-buffer-like class, `SpillableBuffer`, that will _spill_ its device memory to host memory if RMM is...