GMIT Software Development Module Emerging Technologies Original lab problems can be found here
In this lab I am demonstraiting how I use Python to read MNIST data sets and ability to interprit it in some other format, be it a string representation of data to console or a data output to a file e.g., .png.
MNIST data sets can be found here
-
Unzip data set files, read data from a byte stream and place it into appropriate data structures.
-
Study, split and output values from this data structures into meaningful form i.e., convert collection of bytes into an image this bytes represent.
-
Learn new Python techniques and practices used in everyday programming e.g., use of common libraries, separation of modular concerns...
-
runner.py - an entry point to an application. This is the file to run with compiler. It provides a user with a menu to operate application functions.
-
labels_reader.py - module that is responsible for reading labels data sets and returning a list of labels. Each label digit corresponds to a number represented in the image file of the image data set. Those labels are later used to name an image files.
-
images_reader.py - this module reads data sets that store bytes of images and constructs a more complecated data structures of nested lists to be used later.
-
image_saver.py - module with a methods to handle saving of each individual image from a data structure of bytes to a file, giving it appropriate name and file extension.