About This Hackathon

<h3>Welcome to Week 24 of the Weekly MachineHack Hackathon Series!</h3><p>This week brings an exciting new challenge in the domain of machine learning: E-Commerce Forecasting For Sales task is to develop a machine learning model that predicts the future sales quantity of e-commerce products based on historical sales data. Your solution should help e-commerce businesses optimize stock levels, minimize overstocking, and prevent stockouts.</p><h3>Challenge Details</h3><p>Your task is to develop a to develop a machine learning model that predicts the future sales quantity of e-commerce products based on historical sales data.</p><h3>Participation and Benefits</h3><ul><li>Intermediate Level: This hackathon is ideal for participants with a basic understanding of machine learning and deep learning techniques.</li><li>Community Engagement: Join our dynamic community on <a target="_blank" rel="noopener noreferrer" href="https://t.me/joinchat/NJLxnlWiz9lFnEJU20Sccw">Telegram</a> to share ideas, ask questions, and collaborate with fellow participants.</li><li>Certificates: Every participant will receive a certificate from MachineHack, and winners will earn a spot on the leaderboard.</li><li>Live Walkthrough Session: Attend a live session on 5th March 2024 4PM IST to gain valuable insights and tips for approaching this challenge.</li></ul><h3>Submission and Evaluation</h3><ul><li>Submission Format: Participants must submit their predictions in the format specified in submission.csv.</li><li>Evaluation Metric: Submissions will be evaluated based on the RMSE , measuring how well the model predict sales quantity.</li><li>Leaderboard: Track your progress and aim for the top spot on the leaderboard.</li></ul><h3>Data Description</h3><p>The dataset for this hackathon includes:</p><ul><li>train.csv: Contains historical sales data at levels of categories and brands.</li><li>test.csv: Contains data for testing.</li><li>submission.csv: The format in which your predictions should be submitted.</li></ul><h3>How to Crack This Challenge</h3><p>To tackle this challenge successfully, follow these steps:</p><p><strong>Data Pre-processing</strong></p><ul><li>Handle Missing Values: Identify and impute or remove missing values.</li><li>Normalize/Scale Numerical Data: Standardize numerical features to improve model performance.</li></ul><p><strong>Model Development</strong></p><ul><li>Use machine learning models like Random Forest and XGBoost for demand forecasting.</li><li>Implement LSTM (Long Short-Term Memory) for capturing time-series dependencies.</li><li>Utilize LightGBM for efficient and fast gradient boosting.</li></ul><p><strong>Training and Optimization</strong></p><ul><li>Apply Grid Search or Random Search for hyperparameter tuning.</li><li>Train models using historical data to capture trends, seasonality, and sales patterns.</li><li>Use cross-validation to assess model performance and avoid overfitting.</li></ul><p><strong>Validation and Testing</strong></p><ul><li>Ensure the model generalizes well to unseen data.</li><li>Generate predictions for the test dataset in the required format for submission.</li></ul><p>For our subscribers, a starter notebook will be available to guide you through data pre-processing and basic model building. You can customize and enhance this framework to develop your solution further.</p><h3><strong>Getting Started</strong></h3><ol><li><strong>Register Now:</strong> Make sure to register for the hackathon to stay updated.</li><li><strong>Download the Dataset:</strong> Access the dataset from the MachineHack platform and start working.</li><li><strong>Join the Community:</strong> Interact with fellow participants and mentors via our Telegram group for discussions and support.</li></ol><h3><strong>Support and Resources</strong></h3><p>For any questions or assistance, please reach out to our support team at support@machinehack.com. Stay informed about the latest announcements by subscribing to our newsletter.</p><h3><strong>Happy Hacking and Growing! 🚀</strong></h3>

Key Information

• Category: Hackathon
• Difficulty Level: Intermediate
• Status: Expired
• Start Date: 2025-02-18T18:10:00Z
• End Date: 2025-03-16T23:23:59Z
• Current Participants: 114

Prizes and Awards

Knowledge

Rules and Guidelines

<ul><li>The participants are required to provide the code for the work done.</li><li>The output of the code should match the submission file with the "Best Score" achieved by the participant.</li></ul>

Evaluation Criteria

<p>RMSE</p>

Quick Summary

E-Commerce Forecasting For Sales is a intermediate level hackathon currently expired. It has 114 participants. Prizes include: Knowledge. The event runs from 2025-02-18T18:10:00Z to 2025-03-16T23:23:59Z.Registration is free and open to all skill levels.