Highlights: Hello everyone. In this post, we will demonstrate how to build the Fully Connected Neural Network with a Multilayer perceptron class in Python using PyTorch. This is an illustrative example that will show how a simple Neural Network can provide accurate results if images from the dataset are converted into a vector. We are going to use a fully-connected ReLU Network with three layers, trained to predict the output \(y \) from given input…
Read more

Highlights:  In the field of machine learning, the goal of classification is to use characteristics of an object to identify a class to which that object belongs. To classify the object we can use various types of classifiers. In this post, we are going to talk about one particular type of classifiers called Linear Classifiers that can be used to solve easy image classification problems. To better understand them we will conduct several experiments and…
Read more

Highlights: Hello and welcome to our new blog post. So far, we have mainly worked with very simple datasets. To train our models we used linear regression and logistic regression models and for that purpose, we artificially created very easy toy data sets such as blobs. However, in order to fully utilize the power of Deep Learning and neural networks, we will have to work with the larger dataset. Now it is a great time…
Read more

Highlights: In Machine Learning, a backpropagation algorithm is used to compute the loss for a particular model. The most common starting point is to use the techniques of single-variable calculus and understand how backpropagation works. However, the real challenge is when the inputs are not scalars but of matrices or tensors. In this post [1], we will learn how to deal with inputs like vectors, matrices, and tensors of higher ranks. We will understand how…
Read more