Deep Learning Track WiSe 24/25
Deep Learning Track WiSe 24/25
Deep Learning Track WiSe 24/25
  • Welcome to the Deep Learning Track
  • Setup
  • Learning Material
  • Section 1 - The Math
    • Derivatives and Gradients
    • Vectors, Matrices and Tensors
    • The power of matrix computation
    • Exercise - Matrix Computation
  • Section 2 - The Data
    • PyTorch Datasets and Data Loaders
    • Working with Data Tables
    • Exercise - Loading Data from a CSV file
    • Working with Images
    • Exercise - Image Datasets
    • Working with Text
  • Section 3 - Neural Networks
    • Activation Functions
    • Exercise - Activation Functions
    • Exercise - The Softmax Function
    • The Neuron
    • Two type of applications: Regression and Classification
    • Loss Functions
    • Exercise - Regression Loss Functions
    • Exercise - Classification Loss Functions
    • The Gradient Descent Algorithm
    • Exercise - Implementing Gradient Descent
    • Exercise - PyTorch Autograd
    • Exercise - Regression with Neural Networks
    • Exercise - Classification with Neural Networks
    • Playground - Neural Networks
  • Section 4 - Convolutional Neural Networks
    • Convolution
    • Convolutional Neural Networks
    • Classifying handwritten digits
    • Playground - Convolutional Neural Networks
    • Transfer Learning
  • Final Project - Text Classification
  • Further Resources
    • Computer Vision Libraries
    • Image Classification with PyTorch
    • Object Detection with PyTorch
    • Deep AI Explainability
Powered by GitBook
On this page

Section 4 - Convolutional Neural Networks

In this chapter we will have a look at how neural networks can deal with images. The networks we need for this task are called Convolutional Neural Networks.

So far we have covered neural networks that can take tabular data in form of vectors and classify them or build a regression function that tries to predict a true function. But what if our data actually comes in form of images? Can we use the classical neural networks we have seen before that we used for classifying CSV data files or performing regression based on data points? Theoretically we could do this, but those models actually perform bad on image data, because they are bad at finding patterns in 2D or 3D datasets. But there s a solution to that problem. We will introduce a new building block called the "convolutional layer" which will replace our linear layers that we used in previous exercises. The following exercised will help you to understand what happens in convolutional layers and why they can handle image data so well.

PreviousPlayground - Neural NetworksNextConvolution