An implementation of the TensorFlow Tutorial for BasicClassification using the MNIST Fashion Dataset. The original tutorial can be found here.
You can choose from multiple debugging and visualization options which fulfill the purpose of learning and understanding what happens after some important steps.
debug_plot_exampleImage = False # Plots an example image
debug_plot_trainingdata = False # Plots the first 25 images from the training data
debug_print_exampleNormalization = False # Prints an example before and after normalization to console
debug_plot_prediction = False # Plots the image and the predictions per class/category
debug_plot_prediction_imgIndex = 1 # The index of the image which should be used for prediction visualization
This option will plot the first image of the Training Dataset as Colorbar (28x28 pixels with color encoding from 0 to 255)
This option will print the first 25 images in a 5x5 plot.
This option will print the first image data vector before and after normalization to the console.
This option will plot the image (index of the image in the Test Dataset defined by debug_plot_prediction_imgIndex) and the predicted probabilities for each of the 10 classes/categories.
The prediction text will be green if the expected class equals the actual class and red if they differ.
The text below the image
