# Ethos × Genuity — Synthetic Time-Series Challenge

Build realistic synthetic time-series using Genuity’s open-source library train on partial real data and prove your synthetic data generalizes to hidden real samples.

## Welcome & goal
Join **ETHOS :Coding Club, IIT Guwahati** and **Genuity** for a hands-on Kaggle challenge. You will receive **`real_0.6.csv`**, which contains 60% of a multivariate time-series dataset (1,993 rows). The remaining 40% of the real dataset is hidden and will be used exclusively for scoring.  

Your task is to generate a **synthetic dataset matching the full 100% size of the real dataset** (3322 rows). Timestamps need not be preserved, but the timestamp column in your submission must be in strictly ascending order.

## What to submit
- A **CSV file only** that:
  - Matches the schema of `real_0.6.csv`.
  - Contains exactly the **full dataset row count** (3322).
  - Has an extra first column named **`row_id_column_name`**, which must contain serial numbers for each row starting from 0(0 for 1st row and 3321 for last).
- Mandatory to remove 'Series' column.
- Submissions that do not follow these requirements (row count mismatch, wrong column order/naming, or wrong format) will not be scored.
- Include reproducible code in a Kaggle Notebook (baseline notebook and starter kit are provided).

## Why participate
- Gain practical experience with production-grade synthetic data tools.  
- Get an exciting chance to get hired at **Genuity IO**.  
- Compete for prizes, receive mentorship from Genuity IO, and showcase your approach to a wider audience.

## Quick note / rules
- Submissions that trivially copy training rows will be rejected.  
- External public data is allowed if disclosed; no private or proprietary external data.  
- **Scores are calculated across the entire 100% dataset**, with the hidden 40% ensuring fairness and preventing overfitting.


## Github Repo of the Genuity library 
- Github Repo for the library: https://github.com/S-G-mathematics/genuity_os
Mandatory to join this WhatsApp Group
https://chat.whatsapp.com/HBFItT0YtdkAqDaomVfXxM