What's new and planned for Synapse Data Science in Microsoft Fabric
Important
The release plans describe functionality that may or may not have been released yet. The delivery timelines and projected functionality may change or may not ship. Refer to Microsoft policy for more information.
Synapse Data Science provides data scientists with an end-to-end workflow for building their machine learning models, from exploration to model scoring. From a data exploration perspective, data scientists can use R and Python in notebooks, and built-in tools like Data Wrangler for easy analysis. Users can track and compare their model experiments and runs with MLFlow. They can save the best performing model in the workspace as a new model item and easily use Predict for batch scoring at scale. Data science in Fabric is deeply integrated with the rest of the stack, meaning it's seamless to score data in a lakehouse, write back the predictions to OneLake, and visualize the data in reports using Direct Lake mode.
Investment areas
| Feature | Estimated release timeline |
|---|---|
| AI operations in Data Wrangler | Q4 2024 |
AI operations in Data Wrangler
Estimated release timeline: Q4 2024
Release Type: Public preview
A new suite of AI-powered operations in Data Wrangler will allow users to describe code transformations with natural language and generate the corresponding Python; translate custom Python code into PySpark code; and apply SynapseML transformations like text translation and sentiment analysis in a matter of clicks.