-
High levels of PQ compression have allowed us to shrink indexes to minuscule sizes and improve throughput while using refinement to make up for the accuracy loss from the compression.
The problem wi…
-
With the changes in https://github.com/rapidsai/cuml/pull/5293 - we can have hdbscan use all of the distance metrics we support in the raft pairwise distance api, instead of just the euclidean distan…
-
**Is your feature request related to a problem? Please describe.**
Cuspatial example codes currently isn't run in CI and will likely fall out of sync easily. We should execute the example code in CI.…
-
Hashing categorical features to a fixed number of bins is a common preprocessing operation, particularly for tabular deep learning models where memory requirements scale with the number of bins. For e…
-
**Is your feature request related to a problem? Please describe.**
We need to look at Next-generation file formats (NGFF) (https://ngff.openmicroscopy.org/latest/) which use [Zarr](https://zarr.rea…
-
**Is your feature request related to a problem? Please describe.**
Today computing correlation for a dataframe with missing values is not possible, but pandas can do it:
```python
In [10]: gdf = cu…
-
**Is your feature request related to a problem? Please describe.**
There is an issue in spark-rapids that needs the JNI of `replace_nans` in cuDF: https://github.com/NVIDIA/spark-rapids/issues/6276
…
-
### Is there an existing issue for this?
- [X] I have searched the existing issues
### Problem description
For me, all the icons introduced with #14011 are fine, apart from COM which is just the ic…
-
From @tfeher:
Currently everything is [upconverted to fp32](https://github.com/rapidsai/raft/blob/2f08bdd75286688161ca1c8ff3b608985be117bf/cpp/include/raft/neighbors/detail/ivf_pq_build.cuh#L1596).…
-
Thanks for your excellent work!
I recently saw this article on arxiv and would like to ask you for some details.
1. At the "Background Transfer" part. there has a sentence like:
> We then aggrega…