Skip to content

Capstone project for Udacity Machine Learning Engineer Nanodegree program

Notifications You must be signed in to change notification settings

bqnguyen94/Udacity-MLND-Capstone

Repository files navigation

MLND-Capstone

My capstone project for Udacity's Machine Learning Nanodegree

Topic: TalkingData AdTracking Fraud Detection Challenge

Dataset

The training and testing datasets for the Fraud Detection Challenge can be downloaded from Kaggle's competition webpage.

Note: The notebook assumes data files are stored in ./data/ directory.

Requirements

  • Python >= 3.6
  • numpy >= 1.14.3
  • pandas >= 0.23.0
  • scikit-learn >= 0.19.1
  • xgboost == 0.72

Note: It is recommended to use XGBoost with GPU support for better performance. To enable GPU support for XGBoost, please build the library from source with flag -DUSE_CUDA=ON. See more here.

About

Capstone project for Udacity Machine Learning Engineer Nanodegree program

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published