(T) Google has recently announced its new Cloud Machine Learning offering. Google Cloud Machine Learning leverages many Google products: TensorFlow, Dataflow, BigQuery, and Storage. The initial machine learning Python APIs are limited to image analysis, speech recognition, and translation.

TensorFlow was Google-internal research environment for machine learning and deep neural networks that Google open sourced last year. TensorFlow has been used by Google for many of its products: speech recognition in the Google app, smart reply in Gmail, and search in Google Photos. The goal of TensorFlow is now to be both an environment for research and production so that by leveraging the same environment, new research techniques can quickly become integrated into new product developments.

TensorFlow provides numerical computation using data flow graphs:

“Data flow graphs describe mathematical computation with a directed graph of nodes and edges. Nodes typically implement mathematical operations, but can also represent endpoints to feed in data, push out results, or read/write persistent variables. Edges describe the input/output relationships between nodes. These data edges carry dynamically-sized multidimensional data arrays or tensors. The flow of tensors through the graph is where TensorFlow gets its name. Nodes are assigned to computational devices and execute asynchronously and in parallel once all the tensors on their incoming edges become available.”

Part of the Google Cloud Machine Learning offering includes also Dataflow and BigQuery.

Dataflow provides a programming model and a managed service for batch and streaming data processing.

BigQuery is a managed data warehouse. Data is analyzed using common SQL queries.

Reference

TensorFlow – Large-Scale Machine Learning on Heterogeneous Distributed Systems: Whitepaper_tensorflow_2015

Note: The picture above is the TensorFlow logo.

Copyright © 2005-2016 by Serge-Paul Carrasco. All rights reserved.
Contact Us: asvinsider at gmail dot com.