Using Neural Networks to recognize handwritten digits (from 0 to 9).
The mapping of numbers are as follows:
1->1
2->2
3->3
4->4
5->5
6->6
7->7
8->8
9->9
** 0->10
The program first loads a 100 random digits from data.
For instance, this is one of the random actual 100 handwritten examples:
The Neural Network algorithm learns from all examples and predicts the handwritten numbers as:
With a 98% accuracy.