Building Machine Learning Projects with TensorFlow by Rodolfo Bonnin

Key Features

Book Description

This ebook of initiatives highlights how TensorFlow can be utilized in several situations - this comprises tasks for education versions, desktop studying, deep studying, and dealing with quite a few neural networks. each one undertaking presents interesting and insightful routines that may train you the way to take advantage of TensorFlow and convey you ways layers of information might be explored by means of operating with Tensors. easily choose a venture that's according to your surroundings and get stacks of data on the way to enforce TensorFlow in production.

What you are going to learn

• Load, have interaction, dissect, strategy, and retailer advanced datasets
• Solve category and regression difficulties utilizing state-of-the-art options
• Predict the end result of an easy time sequence utilizing Linear Regression modeling
• Use a Logistic Regression scheme to foretell the longer term results of a time series
• Classify photographs utilizing deep neural community schemes
• Tag a suite of pictures and notice positive factors utilizing a deep neural community, together with a Convolutional Neural community (CNN) layer
• Resolve personality popularity difficulties utilizing the Recurrent Neural community (RNN) model

About the Author

Rodolfo Bonnin is a platforms engineer and PhD scholar at Universidad Tecnologica Nacional, Argentina. He additionally pursued parallel programming and photograph realizing postgraduate classes at Uni Stuttgart, Germany.

He has performed examine on excessive functionality computing considering the fact that 2005 and started learning and imposing convolutional neural networks in 2008,writing a CPU and GPU - helping neural community feed ahead degree. extra lately he is been operating within the box of fraud development detection with Neural Networks, and is at the moment engaged on sign type utilizing ML techniques.

Table of Contents

1. Exploring and reworking Data
2. Clustering
3. Linear Regression
4. Logistic Regression
5. Simple FeedForward Neural Networks
6. Convolutional Neural Networks
7. Recurrent Neural Networks and LSTM
8. Deep Neural Networks
9. Running versions at Scale – GPU and Serving
10. Library set up and extra Tips