Abstract
TensorFlow is an open-source software Python-based library developed by Google. It has high popularity in machine learning and deep learning area due to its simplicity, flexibility, and compatibility. In this chapter, we introduce the basic syntax of the TensorFlow and its main operations required to construct an artificial neural network. We briefly introduce the codes for building a recurrent neural network and convolutional neural network for example of MNIST based handwritten digits classification problem.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsReferences
(2018). https://www.tensorflow.org/api_docs/java/reference/org/tensorflow/Session
(2018). https://www.tensorflow.org/tutorials/keras/basic_classification/
LeCun Y (2018) Lenet-5, convolutional neural networks. http://yann.lecun.com/exdb/lenet/
Mnist with CNN code (2018). https://github.com/nlintz/TensorFlow-Tutorials
Hochreiter S, Schmidhuber J (1997) Long short-term memory. Neural Comput 9(8):1735–1780. https://doi.org/10.1162/neco.1997.9.8.1735
Damien A (2018) A recurrent neural network (LSTM) implementation example using tensorflow library. https://github.com/aymericdamien/TensorFlow-Examples/blob/master/examples/3_NeuralNetworks/recurrent_network.py
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Chapter Highlights
Chapter Highlights
-
The TensorFlow is an open-source library that widely used in machine learning and deep learning networks design.
-
A tensor is an R-dimensional array.
-
The Session class allows to evaluate tensors and execute Operation objects
-
add(), multiply() and matmul() are mathematical operations used to add, multiply and perform matrix multiplication of tensors respectively.
-
The connection of a tensor object and an operation object are defined as a graph.
-
As soon as a model is ready, a loss function, optimizer, and metrics should be introduced into the network.
-
Initialize variables before a compilation of the session.
-
Convolutional neural networks (CNN) are widely used in areas related to image classification and recognition.
-
A recurrent neural network is the type of neural network used to solve a time sequence related problems such as a natural language, sound recognition or stock price prediction.
-
The main difference of RNN from a CNN, it sends its output back to the input.
Rights and permissions
Copyright information
© 2020 Springer Nature Switzerland AG
About this chapter
Cite this chapter
Toleubay, Y., James, A.P. (2020). Getting Started with TensorFlow Deep Learning. In: James, A. (eds) Deep Learning Classifiers with Memristive Networks. Modeling and Optimization in Science and Technologies, vol 14. Springer, Cham. https://doi.org/10.1007/978-3-030-14524-8_4
Download citation
DOI: https://doi.org/10.1007/978-3-030-14524-8_4
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-14522-4
Online ISBN: 978-3-030-14524-8
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)