| page_type | languages | products | description | |||
|---|---|---|---|---|---|---|
sample |
|
|
Using Azure ML Tables (MLTable). |
Azure ML Tables (mltable type) allow you to define how you want to load your data files into memory as a Pandas and/or Spark data frame. Azure ML Tables are specific to loading data for ML tasks - such as encodings, type conversion, extracting data from paths, subsetting, etc.
For more information on Azure ML Tables, read Working with tables in Azure ML.
| Notebook | Description |
|---|---|
| Azure ML Tables Quickstart | Demonstrates an end-to-end example of using MLTable, including asset creation, loading into both interactive sessions and jobs. The data is in parquet format. |
| Azure ML Tables Local-to-Cloud | Demonstrates how to work with data and tables locally and upload to the cloud as a data asset for improved sharing and reproducibility. |
| Create an Azure ML Table from Delimited Text Files (CSV) | Demonstrates creating an MLTable from delimited files (CSV). |
| Create an Azure ML Table from Delta Lake table | Demonstrates creating an MLTable from a data lake table on Azure storage. |
| Create an Azure ML Table of paths | Demonstrates creating a Table of paths on cloud storage that can then be streamed into a Python session. |