This notebook trains a neural network for the classification of handwritten digits. It uses the Modified National Institute of Standards and Technology (MNIST) database, that was created by mixing samples taken from American Sensor board employees and those taken from high school students. The training and test dataset consists of 60,000 and 10,000 samples of handwritten digits in grayscale images, respectively. The notebook uses the Keras API to build and train models using Tensorflow.
tatha04 / mnist-handwritten-digits-classification Goto Github PK
View Code? Open in Web Editor NEWA beginners guide to multi-class classification of the MNIST handwritten digits dataset using Neural Networks.
License: Apache License 2.0