This repository contains the code for the machine learning FP project. It is structured into four subtasks which can be found in the four notebooks.
uv is a python package manager which is simple to install and very fast you can find the installation instructions here: https://docs.astral.sh/uv/getting-started/installation/
After installing uv, open this directory in the terminal and run
uv venv
source .venv/bin/activate
uv pip install -r requirements.txtIf you already have conda, you can open this directory in the terminal and execute
conda env create -f environment.yamlAfter you created your environment, set the python interpreter in VSCode to the python environment you just created.
The usage of large language models (LLMs) can drastically speed up the time it takes to solve the tasks. In general, using LLMs for help is allowed and encouraged, with the only exception being the beginning of part 1 up to 1.3.2, which should be simple enough and hopefully instrumental. However, when using LLMs we expect you to understand 100% of the code you submit and be able to explain and reason about it. Your understanding will be checked during the final discussion after you finished all tasks.
Pytorch: You will use pytorch to train a neural network. To get familiar with the package, you can read this pytorch introduction.
K-Nearest Neighbors: geeks4geeks post
PCA: stack exchange explanation, wikipedia
T-SNE: blogpost
UMAP: blogpost
Decision Trees, Random Forests, and Gradient Boosting: blogpost, more details on wikipedia:
Neural Networks: visual series on neural networks, more detailed explanation of convolutional neural networks