24.07.2016 Views

www.allitebooks.com

Learning%20Data%20Mining%20with%20Python

Learning%20Data%20Mining%20with%20Python

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

Chapter 11<br />

Summary<br />

In this chapter, we looked at using deep neural networks, specifically convolution<br />

networks, in order to perform <strong>com</strong>puter vision. We did this through the Lasagne<br />

and nolearn packages, which work off Theano. The networks were relatively easy<br />

to build with nolearn's helper functions.<br />

The convolution networks were designed for <strong>com</strong>puter vision, so it shouldn't<br />

be a surprise that the result was quite accurate. The final result shows that<br />

<strong>com</strong>puter vision is indeed an effective application using today's algorithms<br />

and <strong>com</strong>putational power.<br />

We also used a GPU-enabled virtual machine to drastically speed up the<br />

process, by a factor of almost 10 for my machine. If you need extra power to run<br />

some of these algorithms, virtual machines by cloud providers can be an effective<br />

way to do this (usually for less than a dollar per hour)—just remember to turn<br />

them off when you are done!<br />

This chapter's focus was on a very <strong>com</strong>plex algorithm. Convolution networks take<br />

a long time to train and have many parameters to train. Ultimately, the size of the<br />

data was small in <strong>com</strong>parison; although it was a large dataset, we can load it all in<br />

memory without even using sparse matrices. In the next chapter, we go for a much<br />

simpler algorithm, but a much, much larger dataset that can't fit in memory. This<br />

is the basis of Big Data and it underpins applications of data mining in many large<br />

industries such as mining and social networks.<br />

[ 269 ]

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!