Using Pytorch basics I build and trained a complete neural network. The type of network I build is called a Multilayer Perceptron (MLP), which is fantastic for tabular data.
The project is to predict the quality of wine based on features like alcohol content and density, using a publically available dataset. The target variable of "quality" is a subjective measure of the wine's quality based on expert tasters. I used regression technqiue in this project.
I performed a number of steps:
Acquiring and exploring the dataset Splitting the dataset into train / validation / test partitions Implementing a neural network and optimiser Writing a training loop Visualising results