/CS-189-Machine-Learning

This is the coding homework of cs 189.

Primary LanguageJupyter Notebook

CS 189 Machine Learning

This is my implementation of the coding homework in CS 189.

HW1 Support Vector Machines

  1. Used SVM to classify the MNIST dataset, spam dataset and CIFAR-10 dataset.

  2. Finished hyperparameter tuning and used k-fold cross-validation to validate the model.

HW3 LDA and QDA

  1. Plotted the isocontours of several functions.
  2. Computed eigenvectors of the Gaussian Covariance Matrix
  3. Utilized LDA and QDA to classify the digits and spam dataset

HW4 Logistic Regression

  1. Implemented logistic regression with L2 regularization with gradient decent and stochastic gradient descent.
  2. Varied the learning rate of stochastic gradient descent.

HW5 Decision Trees for Classification

  1. Designed a decision tree and random forest.
  2. Utilized the tree model to classify the spam and Titanic dataset.

HW6 Neural Network

  1. Visualized the MDS189 dataset.
  2. Implemented a muti-layers fully connected network.
  3. Constructed a CNN to complete the classification task.

HW7 Low-Rank Approximation

  1. Used Singular Value Decomposition (SVD) to get the rank-k approximation.
  2. Ran the approach on two different images with different ranks.