dask / dask/dask-ml

Nearest Neighbors

Open
#982 0 comments 3 reactions 0 assignees View on GitHub
Dominant language
Python
Stars
951
Forks
262
PR merge metrics
No merged PRs in 30d

Description

I would like to see out of core nearest neighbor algorithms like those available in [`cuml`](https://docs.rapids.ai/api/cuml/stable/api/#id37) available in `dask-ml`.

I would hope this could be a matter of wrapping an existing out-of-core approximate neighbors library and handling an input `dask.array`.

Contributor guide

Open the contributing guide

Research direction

Start by reviewing the cuML nearest-neighbor API linked in the issue and the dask.array interface mentioned there. Determine which out-of-core approximate-neighbors library could be wrapped and how a Dask array would be handled; done means a defined integration with working nearest-neighbor behavior and tests.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
machine-learning
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.