Train, optimize and explain time series forecasting models

Show the Community!
1

Hello Streamlit Community!

I have recently built an app to visually train, optimize and explain time series forecasting models. It can be helpful to get a baseline for your time series forecasting project, or to explain the decision-making process of a predictive model to a business audience.

Here is the app:
https://share.streamlit.io/maximelutel/streamlit_prophet/main/streamlit_prophet/app/dashboard.py

A Medium article outlining its main features:

The GitHub repository with the source code:

I would be glad to hear your feedbacks!
Have a great day :slight_smile:

2

Thanks for sharing @MaximeLutel, and welcome to the Streamlit community! :balloon:

This is, IMHO, the best implementation of Facebook Prophet I’ve seen in Streamlit. Very nicely done indeed! :raised_hands:

Charly

3

Thanks a lot @Charly_Wargnier !

4

Very detailed implementation @MaximeLutel. Good work.

5

Hi Maxime. This is a superb application and very easy to use. I installed it this morning and have already made discoveries (unfortunately data issues!) with it. Thank you so much for making getting into Prophet so simple!

6

Thank you very much, I’m glad you find it useful!