Defining a convolutional neural network to work on standard data sets for basic computer vision applications
Here we use the Tensorflow library to define a deep neural network for basic computer vision applications. Learn more about Tensorflow.
We define multiple convolutional layers in our model use pooling to reduce the amount of pixel data. We also use callbacks to call off training when our desired accuracy is achieved. We define a separate class for the same. Using Convolutions we can improve the accuracy of our model on both the training and the test set. To learn more about convolutions: Please refer this playlist. Here we will be using the Fashion_MNIST Data set.
Initially the code is meant to be run in google colab notebook environment. It can also be run on your machine either with minor or no changes.
Please feel free to provide feedback, suggestions and commits. Thank You.