Build, deploy, and operate Enterprise AI applications.
Data Integration
Data Connectors
Unified Virtual Data Lake
Unified Data Image
Object Models
Custom
Pre-Built / Extended
APIs
Data Exploration
- Access up-to-date data from a unified data model across the organization. Loaded and virtualized data sources are available without writing complex joins, regardless of the persistence technology.​
- Quickly access and explore raw data files uploaded to a Notebook or file system like Azure Blob storage, AWS S3, or on-premise SSD.​
- Use pre-build no-code interfaces for viewing data, or code your own visualizations and data quality checks, productize your work as a reusable API that the whole team can access from Jupyter or any client.
Visualizations
- Identify patterns and trends by visualizing data in tables and charts.​
- Generate insights by profiling data with histograms, identifying relationships with correlation matrices, and looking at scatter plots, line charts, bar charts, and area charts.​
Jupyter
- Use Jupyter Notebook to view data stored in a unified data model.​
- Easily fetch instances of a BHC3 Type, and view as tables in a notebook.​
- Readily visualize and plot time series data.​
- Use your favorite libraries to visualize and explore data.
Datasets
- Accelerate machine learning prototyping through high- performance, distributed, and in-memory exploration of large datasets from enterprise data sources.
- Auto-record selective data science experimentation steps and metadata for seamless AI application development and deployment.
- Enhance loading, organizing and profiling for unstructured and mixed data including images and text.​
Next
Feature Engineering