1. (20 points) k-nearest neighbors. Complete the k-nearest neighbors Jupyter notebook. The goal of this workbook is to give you experience with the CIFAR-10 dataset, training and evaluating a simple classi er, and k-fold cross validation. In the Jupyter notebook, we’ll be using the CIFAR-10 dataset. Acquire this dataset by running:
wget http://www.cs.toronto.edu/ kriz/cifar-10-python.tar.gz tar -xzvf cifar-10-python.tar.gz rm cifar-10-python.tar.gz
If you don’t have wget you can simply go to: https://www.cs.toronto.edu/~kriz/cifar.
html and download it manually.
We have attached a screenshot of the paths the les ought to be in, in case helpful (though it should be apparent from the Jupyter notebook).
Print out the entire workbook and related code sections in knn.py, then submit them as a pdf to gradescope.
2. (40 points) Support vector machine. Complete the SVM Jupyter notebook. Print out the entire workbook and related code sections in svm.py, then submit them as a pdf to gradescope.
1
3. (40 points) Softmax classi er. Complete the Softmax Jupyter notebook. Print out the entire workbook and related code sections in softmax.py, then submit them as a pdf to gradescope.
2
