platforml is a feature rich Analytics platform enabling:
- Access to big data assets including various Hadoop distributions, processing engines such as Apache Spark and traditional data sources including RDBMS, file servers and Amazon S3 etc.
- Ingestion of data to and from Big Data Repositories
- Perform analytics with custom applications on Hadoop and Spark
- Notebook Integration with private data sources for explorative analytics
- Model Deployment
The following modules are available in the platforml:
Administration of Clusters, Processing Engines and other data sources (Connectors)
Various traditional and big data connectors are available to define in this module which would later be utilized during model development to access directly into jupyter notebooks.
An ingestion module is available to move data to and fro from the data sources defined in the connectors module.
The workbench is where the data scientists and model developers will write programs and conduct exploratory work on model development. Each user on the platform will be able to create project workspaces to define the assets for the models including data files, images, documents, and program code.The workbench also includes a fully functional jupyter notebook integration into the platform to write programs and seamlessly integrate internal data (defined in the connectors) to be available in the notebook.
This module allows the model developer to deploy the trained model to a production-like environment to test real-time scenarios with production data. The model will be automatically exposed as a Restful API for other consumer applications.
For Product demo please contact us at
- Fast Iterative development
- Identify best fit models quickly
- Easy access to your corporate private and cloud data sources
- Model build to Deployment in matter of hours and not days