/rosetta_recsys2019

The 5th Place Solution to the ACM Recsys Challenge 2019 by Rosetta.ai

Primary LanguagePythonApache License 2.0Apache-2.0

The 5th Place Solution to The 2019 ACM RecSys Challenge

Team members

Kung-hsiang (Steeve), Huang (Rosetta.ai); Yi-fu, Fu; Yi-ting, Lee; Zong-han, Lee; Yao-chun, Jan (National Taiwan University); Yi-hui, Lee (University of Texas at Dallas)

Contact: steeve@rosetta.ai

Introduction

This repository contains RosettaAI's approach to the 2019 ACM Recys Challenge. Instead of treating it as a ranking problem, we use Binary Cross Entropy as our loss function. Three different models were implemented:

  1. Neural Networks (based on DeepFM)
  2. LightGBM
  3. XGBoost

Environment

  • Ubuntu 16.04
  • CUDA 9.0
  • Python==3.6.8
  • Numpy==1.16
  • Pandas==0.24.2
  • PyTorch==1.1.0
  • Sklearn==0.21.2
  • Scipy==1.3.0
  • LightGBM==2.2.4
  • XGBoost == 0.9