Welcome to our repository for the 2024 Kaggle Playground Series focused on predicting the age of abalone from various physical measurements. This competition continues the spirit of approachable machine learning challenges, allowing participants to practice and enhance their skills with a fascinating dataset.
Competition Details Objective: Predict the integer age of abalone (measured through the number of rings) using physical measurements. Evaluation: Submissions are evaluated on Root Mean Squared Logarithmic Error (RMSLE). Dataset: The provided datasets (train.csv and test.csv) were generated using a deep learning model trained on the original Abalone dataset. The distributions of features are similar but not identical, providing a unique challenge for competitors. Participants are encouraged to use the original dataset alongside the provided competition data to potentially improve model performance.
Repository Structure
train.csv: Training dataset with the target variable 'Rings'.
test.csv: Test dataset; the goal is to predict the 'Rings' for each entry.
sample_submission.csv: A sample submission file in the correct format.
Regression_with_an_Abalone_Dataset.ipynb.ipynb: Jupyter notebook with model training and prediction steps.
Getting Started
To get started with this competition, you can follow these steps:
Clone this repository to your local machine or Kaggle account. Train your model using model.ipynb and make predictions on the test set. Submit your predictions to Kaggle to see where you stand on the competition leaderboard!
For more detailed instructions and competition rules, visit the competition page.
Contributors This repository is maintained by Sathya Narayanan, who is participating in the Kaggle Playground Series. Contributions, issues, and feature requests are welcome!
Good luck and happy modeling!