sryza / spark-timeseries

A library for time series analysis on Apache Spark
Apache License 2.0
1.19k stars 424 forks source link

Feature/knn #200

Open debasish83 opened 6 years ago

debasish83 commented 6 years ago

@sryza We would like to contribute KNN model to the spark-ts package. Opening it for initial review. I will introduce KNNModel and implement the prediction/forecast using it. Also the lag function should use TimeSeries.lag..we optimized it for performance.

Reference: https://github.com/scikit-learn/scikit-learn/blob/master/sklearn/neighbors/regression.py

debasish83 commented 6 years ago

Closes #201

sryza commented 6 years ago

Hi @debasish83, thanks for your contribution! I left a few comments inline. One broader comment: are you able to scaladoc for the main public classes and methods?