Skip to content

Latest commit

 

History

History
30 lines (24 loc) · 770 Bytes

README.md

File metadata and controls

30 lines (24 loc) · 770 Bytes

Open GPU Data Science

Introduction

This repo contains 8th place Rapids.ai solution of PLAsTiCC Astronomical Classification.

Setup

Requirement

  1. cuda>=9.2
  2. anaconda

Install depencencies

$ conda create -n cudf-latest python=3.6
$ source activate cudf-latest
$ conda install -c nvidia -c rapidsai -c numba -c conda-forge -c defaults cudf=0.4.0
$ pip install xgboost
$ pip install scikit-learn
$ conda install jupyter notebook

Rapids demo notebook

$ cd notebooks
$ jupyter notebook
$ rapids_lsst_demo.ipynb

to be continued.