Summary

PyTorch is the most popular framework for high-performance numerical computation across a wide range of CPU and GPU hardware. While originally intended for quick development of Deep Learning applications, PyTorch is capable of much more and easily applicable to a wide range of numerical applications. In this course, you learn how to leverage the ease of use and power of PyTorch for all your Machine Learning needs. We introduce classical ML and Deep Learning algorithms, analyze their pros and cons, explore the best way to evaluate their performance, and practice how to implement them using PyTorch through a number of practical examples designed to best get you up and running quickly.


Program

  • Machine Learning Overview

    • Basic Principles

    • Unsupervised Learning

    • Supervised Learning

    • Self-Supervised Learning

    • Machine Learning Pipelines

    • Machine Learning as Optimization

    • PyTorch Overview

  • Unsupervised Learning

    • Use Cases

    • Data Preparation

    • Principal Component Analysis

    • K-Means

    • DBScan

    • Latent Dirichlet Analysis (LDA)

  • Supervised Learning

    • Linear Regression

    • Logistic Regression (Classification)

    • K-Nearest Neighbors

    • Support Vector Machines

  • Neural Networks

    • Perceptrons

    • Activation Functions

    • Feed Forward Networks

    • Deep Learning

  • Deep Learning Applications

    • Convolutional Neural Networks

    • Generative Adversarial Networks

    • Recurrent Neural Networks

    • Graph Neural Networks


Resources