fan-loewe/tymon

An AI Assistant More Than a Toolkit

tymon

An AI Assistant More Than a Toolkit The reason for creating framework tymon is simple. making AI more like an assistant, helping us to complete daily AI tasks we want to accomplish, rather than just a tool. I prefer it to be interactive and "human".

Installation

pip install tymon

Example

Timeseries Assistant

prediction task with LSTM

instant a timeseries assistant object, and choose model, set datapath.

from tymon.assistant import TimeSeries
assistant_object = TimeSeries(model_name='LSTM',data_path='./international-airline-passengers.csv')
assistant_object.run() 

run the code you will get a window to set the parameters for model.

set parameters to what you want, click the start button to train the model.

you will get final model in ./ and its performance image.

Related Blog

基于tymon，无需搭建LSTM，航班人数预测