Module 8 - python part 2- Mastering NeuralProphet in Python | Full Walkthrough + Prophet Benchmark

In this module, we explore Prophet and its extension NeuralProphet — powerful forecasting tools designed to tackle real-world business time series challenges. Developed by Facebook, Prophet is known for its ease of use, interpretability, and flexibility in modeling complex components like holidays, multiple seasonalities, and growth constraints. We then expand the scope by introducing NeuralProphet, which builds on Prophet by adding autoregressive modeling and a neural network component to capture short-term dependencies and richer dynamics.

we do this module in 5 parts:
1- Facebook Prophet (Theory),
2- Prophet python basics
3- Prophet python advanced
4- NeuralProphet (Theory)
5- NeuralProphet (Python) (this video)

Lecture timestamps:
0:00 intro and where to find the materials?
4:30 Neural prophet python documentation (installation +) basics
13:16 Neuralprophet - Trend
15:54 Neuralprophet - Seasonality
20:00 Neuralprophet - AutoRegressive
25:17 Neuralprophet - Lagged Regressors
29:07 Neuralprophet - Future regressors and Events
33:51 Other features
41:37 prophet vs neuralprophet performance (Airline passenter data)
43:26 List of hyperparameters in neuralprophet
1:01:55 Uncertainty modeling (quantile regression and naive conformal predictions)
1:11:01 Train-test and cross validation in neural prophet
1:18:01 prophet vs neuralprophet performance (Apple Stock Price data)

Relevant playlists:
Instructor: Pedram Jahangiry

All of the slides and notebooks used in this series are available on my GitHub page, so you can follow along and experiment with the code on your own.

[pedramjahangiry]

🎉 This is the final video of the Deep Forecasting course!
Thanks for sticking with me through this journey — from classical models to NeuralProphet. I hope this series gave you both practical tools and deeper intuition around time series forecasting.