How to make ML models without a single line of code?

Here's a guide on Azure AutoML

/🧵/

AutoML is a feature in Azure that allows you to train and deploy machine learning models with minimal effort and expertise.

Provide the data and specify the task you want to perform (classification or regression) and it will automatically train and tune a machine-learning model.

The process of using AutoML in Azure typically involves the following steps:

1/ Dataset

Upload the data set as a data asset in the Azure ML studio

2/ Preprocessing

Before you can use AutoML, you need to prepare your data.

This typically involves cleaning and formatting the data which will be handled by Azure itself, as well as splitting it into training and test sets which has to be done manually.

3/ Training

Once your data is ready, you can use AutoML to train one or more machine-learning models on it.

AutoML will automatically select the appropriate algorithms and hyperparameters, and it will train and evaluate multiple models in parallel to find the best one.

4/ Model Selection

Once the training has been done, Azure will give you an output of all the models trained with the respective parameters for evaluation.

Select the model as the best fit for your use case.

5/ Deployment

After the selection of the trained model, you can deploy it to Azure for use in production.

This involves creating a container image and deploying it to Azure Container Instances or Azure Kubernetes Service.

6/ Monitoring and maintenance

After deploying your model, you will want to monitor its performance to ensure it is functioning as expected.

You may also need to perform ongoing maintenance tasks such as retraining the model as new data becomes available.

That's a wrap!

If you enjoyed this thread:

1. Follow me @freest_man for more of these
2. RT the tweet below to share this thread with your audience