Machine learning is hard. It used to be harder though, and I feel like ML is getting more and more accessible each day. But acquiring the right background needed to understand what’s going on under the hood of PyTorch or scikit-learn or whatever library you happen to be using is, well, still hard. It requires a lot of work, as the brilliant A Super Harsh Guide to Machine Learning likes to remind us!
Using Azure Automated Machine Learning
At a high level, all auto ml needs is some labelled data and a computer to run on, and this can be either your local computer or some machine in the cloud. Something like in the image below.
Once started, Automated ML will use the compute you hand it over to run multiple experiments on your data, trying out various combinations of algorithms & hyper-parameters, until it trains a good-enough model, which you can then use and integrate in whatever app you might be building.
Read more on Vlad's blog, Head of AI at Strongbytes: https://vladiliescu.net/automl-in-azure-getting-started/