You're reading from Python Machine Learning, Second Edition Machine Learning and Deep Learning with Python, scikit-learn, and TensorFlow

Product type Paperback

Published in Sep 2017

Publisher Packt

ISBN-13 9781787125933

Length 622 pages

Edition 2nd Edition

Languages

Python

Tools

Scikit-learn

Concepts

Deep Learning

Authors (2):

Vahid Mirjalili

Sebastian Raschka

View More author details

Table of Contents (18) Chapters

Preface

1. Giving Computers the Ability to Learn from Data

2. Training Simple Machine Learning Algorithms for Classification FREE CHAPTER

3. A Tour of Machine Learning Classifiers Using scikit-learn

4. Building Good Training Sets – Data Preprocessing

5. Compressing Data via Dimensionality Reduction

6. Learning Best Practices for Model Evaluation and Hyperparameter Tuning

7. Combining Different Models for Ensemble Learning

8. Applying Machine Learning to Sentiment Analysis

9. Embedding a Machine Learning Model into a Web Application

10. Predicting Continuous Target Variables with Regression Analysis

11. Working with Unlabeled Data – Clustering Analysis

12. Implementing a Multilayer Artificial Neural Network from Scratch

13. Parallelizing Neural Network Training with TensorFlow

14. Going Deeper – The Mechanics of TensorFlow

15. Classifying Images with Deep Convolutional Neural Networks

16. Modeling Sequential Data Using Recurrent Neural Networks

Index

Training neural networks efficiently with high-level TensorFlow APIs

In this section, we will take a look at two high-level TensorFlow APIs—the Layers API (tensorflow.layers or tf.layers) and the Keras API (tensorflow.contrib.keras).

Keras can be installed as a separate package. It supports Theano or TensorFlow as backend (for more information, refer to the official website of Keras at https://keras.io/).

However, after the release of TensorFlow 1.1.0, Keras has been added to the TensorFlow contrib submodule. It is very likely that the Keras subpackage will be moved outside the experimental contrib submodule and become one of the main TensorFlow submodules soon.

Building multilayer neural networks using TensorFlow's Layers API

To see what neural network training via the tensorflow.layers (tf.layers) high-level API looks like, let's implement a multilayer perceptron to classify the handwritten digits from the MNIST dataset, which we introduced in the previous chapter. The MNIST...