Using UV let's prepare our environment first. UV will install all our dependencies.

uv sync

You can try to install pytorch with a GPU backend, if you have one available.

uv pip install torch --torch-backend=auto

To run the Python Notebooks, we can use the Jupyter Lab. Simply start the lab with

uv run --with jupyter jupyter lab

For this project we will focus on the components we will regularly use while developing and training networks. Take a look in the files in layers/.

You will need to fill the following sections in each file.

##############################
# YOUR CODE HERE

##############################

Here are the explanations of the tasks for this project.

• Linear Layer Implementation

You can test your progress by simply running uv run pytest .\tests\ -v in the homework folder. The goal of this project is to pass all the tests by implementing the necessary parts in our layers\ folder.