Hands-On Unsupervised Learning Using Python by Ankur A. Patel

Author:Ankur A. Patel
Language: eng
Format: epub, pdf
Publisher: O'Reilly Media
Published: 2018-11-04T16:00:00+00:00

TensorFlow

Before we introduce autoencoders, let’s explore TensorFlow, the primary library we will use to build neural networks. TensorFlow is an open source software library for high-performance numerical computation and was initially developed by the Google Brain team for internal Google use. In November 2015, it was released as open source software.2

TensorFlow is available across many operating systems (including Linux, macOS, Windows, Android, and iOS) and can run on multiple CPUs and GPUs, making the software very scalable for fast performance and deployable to most users across desktop, mobile, web, and cloud.

The beauty of TensorFlow is that users can define a neural network—or, more generally, a graph of computations—in Python, and can take the neural network and run it using C++ code, which is much faster than Python.

TensorFlow is also able to parallelize the computations, breaking down the entire series of operations into separate chunks and running them in parallel across multiple CPUs and GPUs. Performance like this is a very important consideration for large-scale machine learning applications like those that Google runs for its core operations such as search.

While there are other open source libraries capable of similar feats, TensorFlow has become the most popular, partly due to Google’s brand.

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.