/ML-assignments

Machine Learning assignments, Machine Learning (IE500618) course, fall 2022.

Primary LanguageJupyter Notebook

ML assignments

This repository contains the mandatory assignments from NTNU's "Machine Learning" (IE500618) course, fall 2022.

These assignments are mandatory but do not count towards the final grade in the subject.

Contents

  • Use the UCI Mushroom data set
  • Use a multilayer perceptron (MLP) classifier.
  • Clean and split the data into training, validation, and testing.
  • Present the results:
    • Plot the accuracy and loss.
    • Create a confusion matrix.
  • Simulate distributed machine learning using ensemble learning and compare it to a monolithic model.
  • Use the MNIST data set
  • Use a multilayer perceptron (MLP) classifier.
  • For the ensemble model:
    • Divide the data into 3 local sections, by digits: 0-2, 3-5, and 5-9.
    • Train each local model with only one of the sections.
    • Aggregate the 3 local models into a single ensemble model.
  • Present the results:
    • Plot the accuracy and loss.
    • Create a confusion matrix.
    • Make comparisons between the ensemble model and the monolithic model trained on the full dataset.
  • Use the ResNet50 model (transfer learning) for classification.
  • Use the CIFAR-100 data set
  • Present the results:
    • Plot the accuracy and loss.
    • Create a confusion matrix.