IMPORTANT: PUT YOUR NETID IN THE FILE netid in the root directory of the assignment.
This is used to put the autograder output into Canvas. Please don't put someone else's netid
here, we will check.
In this assignment, you will:
- Implement Lp-norm regularization
- Implement two types of loss functions
- Do experiments to understand gradient descent, loss landscapes, and regularization.
To clone this repository install GIT on your computer and copy the link of the repository (find above at "Clone or Download") and enter in the command line:
git clone YOUR-LINK
Alternatively, just look at the link in your address bar if you're viewing this README in your submission repository in a browser. Once cloned, cd into the cloned repository. Every assignment has some files that you edit to complete it.
See problems.md for what files you will edit.
Do not edit anything in the tests directory. Files can be added to tests but files that exist already cannot be edited. Modifications to tests will be checked for.
Make a conda environment for this assignment, and then run:
pip install -r requirements.txt
The test cases can be run with:
python -m pytest -s
at the root directory of the assignment repository.
Simply open an issue on the starter code repository for this assignment here. Someone from the teaching staff will get back to you through there!