Skip to content

Latest commit

 

History

History
32 lines (25 loc) · 1.77 KB

README.md

File metadata and controls

32 lines (25 loc) · 1.77 KB
Raw

Fundamentals of Data Science - Homework 2

Overview

This repository contains the second homework assignment for the "Fundamentals of Data Science" course, conducted in the Winter Semester of 2023. The primary focus of this assignment is on logistic regression, multinomial classification, and transfer learning.

Contents

  • HW2.ipynb: A comprehensive Jupyter notebook that includes:

    • Logistic Regression implementation using Gradient Ascent.
    • Analysis and application of logistic regression to both synthetic and real datasets.
    • Multinomial classification techniques.
    • Exploration of transfer learning using pre-trained models on the CIFAR-10 dataset.
    • Detailed instructions and hints are provided throughout the notebook to guide the analysis.

  • Data:

    • data/Invistico_Airline.csv: Dataset used for logistic regression analysis.

Key Features

  • Detailed implementation of logistic regression from scratch.
  • Exploration of various aspects of logistic regression, including log-likelihood, gradient ascent rule, and decision boundaries.
  • Application of multinomial classification and analysis of its performance.
  • A bonus section on transfer learning, demonstrating the use of pre-trained models for classification tasks.

Instructions

The notebook is divided into several sections, each dealing with different aspects of data science methodologies:

  1. Logistic Regression with Gradient Ascent
  2. Multinomial Classification
  3. Transfer Learning on CIFAR-10 (Bonus Section)

Students are encouraged to follow the instructions and hints provided in each section to complete the assignment.

Setup

To use this repository, clone it and ensure you have Jupyter Notebook installed. The required libraries and dependencies are listed within the notebook.