piyushmahatkar/minst-digit-recognition

Digit Recognition using 2 layer convolutional Neural Network on Keras - Tensorflow

MINST-Digit-Recognition

Digit Recognition using 2 layer convolutional Neural Network on Keras - Tensorflow

Overview:

Resulting accuracy on 5th epoch: Training Accuracy: 98.20% Test accuracy: 99.04% Train on 60000 samples, validation on 10000 samples

Future Updates:

Parameter tuning needed. filter sizes need to change, code can be optimized for computation. Soon I will upload a new code, with better amalgam of networks and parameters.

Model Summary

Layer (type)       Output Shape       Param

conv2d_1 (Conv2D)       (None, 22, 22, 100)       5000
conv2d_2 (Conv2D)       (None, 19, 19, 10)       16010
max_pooling2d_1       (MaxPooling2 (None, 9, 9, 10)       0
dropout_1 (Dropout)       (None, 9, 9, 10)       0
flatten_1 (Flatten)       (None, 810)       0
dense_1 (Dense)       (None, 128)       103808
dropout_2 (Dropout)       (None, 128)       0
dense_2 (Dense)       (None, 10)       1290

Total params: 126,108 Trainable params: 126,108 Non-trainable params: 0