Large Scale Machine Learning with Python by 2016

Author:2016
Language: eng
Format: epub
Publisher: Packt Publishing

By now, this method will be quite familiar; it is basically the same way a classifier in Scikit-learn would work. However, there are two important things to notice. With SkFlow, we can use NumPy and TensorFlow objects interchangeably so that we don't have to merge and convert objects in and out of tensor frames. This makes working with TensorFlow through a higher-level method like SkFlow much more flexible. The second thing to notice is that we applied the toarray method to the main data object. This is because the dataset is quite sparse (lots of zero entries), and TensorFlow is not able to process sparse data well.

Neural networks is where TensorFlow excels and in SkFlow, it is quite easy to train a neural network with multiple layers. Let's perform a neural network on the diabetes dataset. This dataset contains diabetes metrics (binary target) diagnostic features of pregnant females of over 21 years of age and of Pima heritage. The Pima Indians of Arizona have the highest reported prevalence of diabetes of any population in the world and therefore this ethnic group has been a voluntary subject of diabetes research. The dataset consists of the following features:

Number of times pregnant

Plasma glucose concentration at two hours in an oral glucose tolerance test

Diastolic blood pressure (mm Hg)

Triceps skin fold thickness (mm)

2-hour serum insulin (mu U/ml)

Body mass index (weight in kg/(height in m)^2)

Diabetes pedigree function

Age (years)

Class variable (0 or 1)

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.