Searching for tutorials and software about Deep Learning and Neural Nets? Be sure to look at my Resource Page!
Looking for Octave? Go to my Easy Octave on Mac page!

Sunday, January 31, 2016

Still working away at the coalface

When you don't know a programming environment, your life consists of using Google to locate code code fragments, mod them try and get them to work, and then try to figure out how they work. 

At some point I guess I'll have learnt some Python, and feel comfortable with Numpy broadcasting and slicing and dicing arrays, and feel comfortable writing my own code without checking the shape() of each array every minute. That day I'll probably be told that Python is obsolete ...

At the moment, my hardest problem with iPython/Jupyter was to find a clean way to inline plots.

Either of the following two lines seems to work ok for now.
%matplotlib inline
%matplotlib notebook

I used a Hinton diagram module to display character images from the in-memory matrix. 

I'm starting to think one might profit from keeping around the original file data before it gets normalized, shuffled. 

Anyway, here is today's proof of work.

No comments:

Post a Comment

Hey, let me know what you think of my blog, and what material I should add!