In this assignement you will implement several Image Classification models. You will use cross-validation to best choose the hyper-parameters for each model.

Setup

Once you downloaded the attachement. You could work either locally or in a google colab.

Wokring with google colab.

If you choose to work with google colab, you need to adapt the first cell to point to your google drive folder.

First Upload the assignement to your drive and unzip-it.
You need to adapt the first cell to point the folder that you created.
Let’s say that we saved our folder as /Courses/DeepLearning/assignment1/, then you need to update the FOLDERNAME variable like.

# This mounts your Google Drive to the Colab VM.
from google.colab import drive
drive.mount('/content/drive')

# TODO: Enter the foldername in your Drive where you have saved the unzipped
# assignment folder, e.g. 'cs321/assignments/assignment1/'
FOLDERNAME = 'Courses/DeepLearning/assignment1'
assert FOLDERNAME is not None, "[!] Enter the foldername."

# Now that we've mounted your Drive, this ensures that
# the Python interpreter of the Colab VM can load
# python files from within it.
import sys
sys.path.append('/content/drive/My Drive/{}'.format(FOLDERNAME))

# This downloads the CIFAR-10 dataset to your Drive
# if it doesn't already exist.
%cd /content/drive/My\ Drive/$FOLDERNAME/cs231/datasets/
!bash get_datasets.sh
%cd /content/drive/My\ Drive/$FOLDERNAME

Working locally.

You could also work locally using your conda setup. In this case, you don’t need to setup any FOlDERNAME since it will be in the current folder. So your first cell will be as:

# This downloads the CIFAR-10 dataset to your Drive
# if it doesn't already exist.
%cd cs231/datasets/
!bash get_datasets.sh
%cd ../../

This will help download the FIFAR-10 Data set

Q1: k-Nearest Neighbor classifier

The notebook knn.ipynb will walk you through implementing the kNN classifier.

Q2: Training a Support Vector Machine

The notebook svm.ipynb will walk you through implementing the SVM classifier.

Q3: Implement a Softmax classifier

The notebook softmax.ipynb will walk you through implementing the Softmax classifier.

Q4: Two-Layer Neural Network

The notebook two_layer_net.ipynb will walk you through the implementation of a two-layer neural network classifier.

Q5: Higher Level Representations: Image Features

The notebook features.ipynb will examine the improvements gained by using higher-level representations as opposed to using raw pixel values.

Assignment #1 - Image Classification using ML and Two-Layer Neural nets

Late Policy