Preprocessing and feature engineering with Azure Machine Learning DataPrep
In this section, we will dive deeper into the preprocessing and feature extraction process using Azure Machine Learning. We will first access and extract data with different data formats from different storage systems, such as text data and CSV data from blob storage, and tabular data from relational database systems.
Then, we will take a look at common data transformation techniques using Azure Machine Learning DataPrep, a Python library to build transformations on top of datasets directly in Azure Machine Learning. You will also learn common techniques of how to filter columns, split columns through expressions, fix missing values, convert data types, and even how to derive transformations through examples.
Finally, we will write the data back into data storage where it can be registered as a cleaned dataset in Azure Machine Learning. By doing this you can implement fully enterprise-grade ETL and data...