Defining regression

Our first task is to define the model which can perform regression on provided MNIST dataset. So we will create a TensorFlow model with 2 hidden layers that are in a Fully Connected Neural Network. You may also hear it referred to as MLP.

The model will perform the operation that will fit the Equation 2.1 where y is the label, x is the image, W is the weight which model will learn and b is the bias which will also be learned by the model:

Equation 2.1: The regression equation for the model

SUPERVISED LEARNING: When you have data and accurate labels for the training set (i.e. you know the answer) you are in a Supervised Deep Learning paradigm. Model training is a mathematical process by which the features of the data are learned and associated with the proper labels so that when a new data point (test data) is presented the accurate output class label can be produced (that is, when you present a new data point and do not have the label (don't know the answer) your model can produce it for you with a highly reliable class prediction.

Each iteration will try to generalize the values of weight and bias and reduce the error rate. Also keep in mind, that we need to ensure that the model is not overfitting which may lead to wrong predictions for the unseen dataset. We'll show you how to code this and visualize the progress to aid in your intuition of model performance.

Table of Contents for Defining regression

Create new playlist

Sign In

Sign Up

Table of Contents for
Defining regression