Creating a data asset using the Python SDK

In this section, we will show you how to create a data asset using the Python SDK. As mentioned in the previous section, you can create data from datastores, local files, and public URLs. The Python script to create a data asset from a local file (for example, titanic.csv) is shown in Figure 2.19.

Please note that in the following code snippet, type = AssetTypes.mltable abstracts the schema definition for the tabular data, making it easier to share datasets:

Figure 2.19 – Creating a data asset via the Python SDK

Inside the my_data folder, there are two files:

• The actual data file, which in this case is titanic.csv
• The mltable file, which is a YAML file specifying the data’s schema so that the mltable engine can use it in order to materialize the data into an in-memory object such as pandas or DASK

Figure 2.20 shows the mltable YAML file for this example:

...