Hand Written Recognition ANN

Hand writing recognition using Artificial Neural Networks (ANN)

dataset “Digit” includes 1797 small images (8x8 pixels), each one includes a hand-written digit (0-9)

ANN Accuracy: 0.916666666667

Grid Best Accuracy Score: 0.961046188091

Grid Best Parameters: {'hidden_layer_sizes': (126,)}

Backpropagation In the following Neural Network, we have initialized the weights randomly. Use a training sample (X,y) = ((1,1), 0) to update the weights (perform one round of backpropagation using one training sample). Use learning rate parameter α = 0.1.

• Build the feature matrix and label vector: Each image is considered as a data sample with pixels as features. Thus, to build the feature table you have to convert each 8x8 image into a row of the feature matrix with 64 feature columns for 64 pixels

• Design and Train an ANN with one hidden layer with 80 neurons to recognize the digits based on the training dataset that you built in part (c). Use random_state=1, learning_rate_init = 0.002. Then, Test your ANN on testing set (from part(c)), and calculate and report the accuracy. Also, calculate and report the Confusion Matrix.

author: yosep kim

github.com/cowboyuniverse