Installation

EasyTPP provides an open-source library for Neural TPP, with a fully automated pipeline for model training and prediction.

Requirements

PyTorch version >= 1.8.0
Python version >= 3.7
Tensorflow version >= 1.13.1 (only needed when using Tensorflow backend)

First, we need a python environment whose version is at least greater than 3.7.0. If you don’t have one, please refer to the Documentation to install and configure the Anaconda environment.

conda create -n easytpp python=3.8
conda activate easytpp

Then, install Pytorch and keep the version at least greater than 1.8.0.

pip install torch

By default, we assume to use PyTorch. If one wants to use Tensorflow backend, please install tensorflow additionally. Both Tensorflow 1.13.1 and 2.x are supported.

pip install tensorflow

Install

Install with pip

pip install easy-tpp

Install with the source

Setup from the source：

git clone https://github.com/ant-research/EasyTemporalPointProcess.git
cd EasyTemporalPointProcess
python setup.py install