ODBC Reader for TensorFlow

The ODBC Reader for TensorFlow allows users to connect Google's TensorFlow machine learning project with any SQL-92 compliant ODBC Driver.

The ODBC Reader is driver-agnostic and can retrieve data from any ODBC Driver source. This includes our ODBC Drivers, as well as open source and other standards-compliant third-party ODBC Drivers.

The ODBC Reader tensorflow-odbc and full source code are publicly available on GitHub: http://github.com/cdatasoftware/tensorflow-odbc/

Connecting TensorFlow to ODBC

Machine learning enables organizations to leverage today's vast computational power to recognize patterns in data and make meaningful predictions. To do this, machine learning platforms have to crunch enormous amounts of data. From financial fraud detection to enabling self-driving car technologies, data is essential to pattern recognition and discovery.

Unfortunately, meaningful organizational data is often siloed into any number of disparate systems. From CRM, ERP, and SaaS applications to various RDBMS or NoSQL database sources, every data interface is different. Organizations interested in applying machine learning techniques are forced to write data-source-specific custom code or live with static dumps to populate their models.

Our goal in this project is to help streamline the way that organizations connect machine learning with real-time data, regardless of location. By doing so we hope to facilitate the growth of ML integration and ultimately make it more accessible to businesses of all sizes.


The ODBC Reader is an open-source project released under the GPL 3.0 license. For information about commercial licensing, please contact our sales team.