How to manage big data with Classification Learner

9 visualizaciones (últimos 30 días)
Francesco Calabrese
Francesco Calabrese el 20 de Abr. de 2018
Respondida: Bernhard Suhm el 22 de Abr. de 2018
Hi everybody, first time writing here. I am working for my thesis regarding the prediction in maintenance and I've seen that Matlab provides the app Classification Learner to forecast. The problem is that I have to manage a huge amount of data (in example 3 columns of 6 mln rows each one) that causes an OutofMemory error. I've also seen that there are Tall Arrays: I've read the tutorial and seen the video but not understood how to use this Tall Arrays with the app? Is there a way to use both these tools? Or can you suggest other solutions? Thanks in advice

Respuesta aceptada

Bernhard Suhm
Bernhard Suhm el 22 de Abr. de 2018

"Tall arrays" are currently not supported from within the Learner apps. If your data doesn't fit into memory, you have to use the programmatic workflow. But many machine learning algorithms do accept tall arrays, and if they do, you can first prototype them with a smaller in-memory dataset, and then simply replace the in-memory training array with a tall array. You must already have found the documentation, but here's a good guide how to use tall arrays with machine learning .

Más respuestas (0)

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by