tsboost 0.1.0

Time series Framework

Context

TSBoost is a framework for time series forecasting.

It mixes classical statistics practices with non linear optimisation techniques of current Machine Learning.

Requirements

32-bit Python is not supported. Please install 64-bit version.

TSBoost uses gradient boosting optimisation provided by XGBoost & LightGBM, both have C++ source code and need a compiler.

For Windows users, VC runtime is needed if Visual Studio (2015 or newer) is not installed.

For Linux users, glibc >= 2.14 is required

sudo apt-get install build-essential # Ubuntu/Debian

sudo yum groupinstall ‘Development Tools’ # CentOS/RHEL

For macOS users, install OpenMP librairy

brew install libomp

Installation

After installing the compiler, install from PyPI Using pip

pip install tsboost

Quick Start

You can get started with a jupyter notebook tutorial : TSBoot quick start

History

0.1.0 (2019-06-10)

• First release on PyPI.