How do I learn multivariate time series forecasting?

This course teaches univariate time series forecasting

The tensorflow developer certificate requires that we can “Train models to predict values for both univariate and multivariate time series”

I imagine this is an addition to the handbook that did not exist when this coursera course was created

What can I do to learn multivariate time series forecasting in tensorflow?

In case anyone is interested, I learnt how to do one form of multivariate time series forecasting by:

  1. attempting to read and understand this notebook: I failed, as it was too complex a level for me right now.

  2. reading this: Time Series Forecasting as Supervised Learning. More specifically this part:

    This was a simple explanation on how to do multivariate forecasting that was directly related to what I learnt for univariate forecasting on TF

  3. then grabbing the dataset from (1) and creating a notebook from scratch using the model and data processing methods I learnt in course 4 to do the Melbourne temperature assignment. This seemed simple at first, with minor changes, but I had to introduce and tweak several things (like scaling, and the model lambda layer for example), but when I managed to get a good forecast it felt like a great accomplishment.

Well, I am also having difficulties with Multivariate forecasting using LSTM, and my data is a sensor data that I fetched from an industrial machinery, though, haven’t yet found a solid solution for preparing the data for LSTM, and still on the look! Thank you though for sharing your experience with us!

1 Like