This repository contains the code for multiple projects related to data analysis and machine learning. Unfortunately, due to the large size of the data used in some of these projects, I was unable to push it to this repository. However, I have provided links to where you can find the data for each project.
This project explores the classification of Osteoporosis in Knee X-rays using Deep-Learning and Transfer Learning Techniques. The code for this project can be found in the osteoporosis directory.
Data The data used in this project can be found at the following link: https://www.kaggle.com/datasets/stevepython/osteoporosis-knee-xray-dataset.
This project involves predicting weather patterns using machine learning techniques.
The WiDS Worldwide team at Stanford University, Harvard University IACS, Arthur, and the WiDS Datathon Committee organize this yearly datathon. This years' competition aims to improve sub-seasonal weather forecasting using machine learning. Accurate long-term forecasts of temperature and precipitation are crucial to help people prepare for extreme weather events caused by climate change.
Data The dataset was created in collaboration with Climate Change AI (CCAI) and can be found link: https://www.kaggle.com/competitions/widsdatathon2023/data.
In this project, I performed Exploratory Data Analysis (EDA) to generate some descriptive insight. The steps taken before the EDA involved
- Importing the data
- Assessing the data content to uncover possible issues
- Fixing issues identified which ranged from duplicated values to datatype issues
- Price trend over time
Theres an upward trend in cars prices with respect to year. Manufacturer A process the most expensive car
- Price by car type
Camry, Hero Super Splendor and 800 are 3 of the least expensive cars to purchase. On the other hand, the most expensive cars include but not limited to Honda, Bajaj, TVS - (This analysis is relative to the dataset used only)
- Distance Travelled vs price and Fuel type
The distance travelled by a car has no relationship to its price
- Present Price by selling price
The present car price and its selling price have a strong inverse linear relationship. This implies that an increase in one variable leads to a decrease in the other
- Kilometer driven by transmission type
There are only 2 types of transmission. And Automatic vehicles have more records, hence more distance travelled
- Car type by transmission type
Most cars produces use Manual transmission
- Car type by fuel type use
The most used fuel type is petrol with a wide margin. Only two types of cars use Compressed Natural Gas (CNG)
- Price by fuel type
Surprisingly, CNG is the most expensive fuel type, undermining its the list used. The cheapest is diesel
- Car type by Manufacturer
Manufacturer A produces the most cars. - Almost looks like a monopoly market for A, while B produces only 9 cars and C produces only 1
- Price by Manufactuerer
Manufacturer does not just produce the most cars, his cars are also the most expensive
