This project is an extension of the 4th lesson from FastAI 2020 course. Here the mnist dataset was used to train a CNN to classify handwritten digits.
The trained model is exported out ("export.pkl") and loaded python script which creates a GUI tool (Tkinter). This tool allows the user to handwrite a digit onto the given canvas and then let the computer predict the number based on the trained model. THe resulting "guess" is shows with a probability (as well as the probability of all digits).
Care must be taken when drawing as to make the digit large and centered. Off center numbers are difficult for the computer to recognize.