Issue: #37 – Dataset Understanding and Exploratory Data Analysis

[Krishna200608]

Issue: #37

Summary

This Pull Request focuses on understanding the SpaceNet Astronomy Image Dataset and performing an initial exploratory data analysis (EDA) before any model building.

The goal of this contribution is to analyze the dataset structure, identify classes, study class distribution, visualize sample images, and examine image properties such as resolution and format.

Work Done

• Explored the dataset directory structure and identified all astronomy object classes
• Computed the number of samples per class and analyzed class balance
• Visualized random image samples from each class
• Analyzed image resolutions and formats across the dataset
• Observed that all images are uniformly sized at 2048 × 2048 pixels

Key Observations

• The dataset consists of 8 classes: asteroid, black hole, comet, constellation, galaxy, nebula, planet, and star
• Class distribution is well-structured and clearly observable
• All images have a standardized resolution, simplifying downstream preprocessing

Implementation Details

• Work was performed using a Kaggle notebook
• The exploratory notebook has been added under the required directory structure:
  participants/IIT2023139/data_exploration.ipynb
• All OpenCode contribution guidelines and PR template rules have been strictly followed

Environment

• OS: Windows 11
• Editor: VS Code
• Platform: Kaggle Notebook

This PR addresses only Issue #37 and does not overlap with any other issues.

[OpenGitBot]

Hey @Krishna200608

Thanks for opening this PR 🚀. Mentor will review your pull request soon and till then, keep contributing and stay calm.

Thanks for contributing in OpenCode'25 ✨✨!