Digital Legends & Spirit Guardians

🌌 Neural Realms & Mystic Entities: Hybrid Data Science & Fantasy RPG Attributes 🐲⚙️📊
About the Dataset
This high-dimensional synthetic dataset (70,000+ records) is a unique fusion of Deep Learning Technical Stacks and Fantasy RPG Mechanics. It was meticulously architected to simulate real-world data challenges, making it the perfect playground for Senior SQL Developers and Data Scientists.

The dataset features a rich mix of 16 attributes, blending coding frameworks (PyTorch, TensorFlow) and architectures (Transformers, CNNs) with mystical traits like Mana Levels, Spirit Animals, and Arcane Realms.

File Descriptions

1. 📂 train.csv
   The primary training set containing 70,000 rows and 16 columns.

Target Variable: Success_Probability (Continuous value between 0 and 1).

Key Features: Tech Stack (Python, PyTorch, etc.), RPG Stats (Mana, Stamina), and Environmental context (Realms).

Data Traps: Includes intentionally injected Outliers in Experience Points, Missing Values in Health Status, and Corrupted Labels (DATA_VOID_ERROR) to test your cleaning skills.

2. 📂 test.csv
   A secondary dataset with 30,000 rows and 15 columns (excluding the Target). Use this to evaluate your model's performance or to practice "Data Drift" analysis between training and testing distributions.

3. 📂 submission.csv
   A sample submission file containing:

Entity_ID: The unique identifier for each character.

Success_Probability: Your predicted success rate.

4. 📂 analysis_queries.sql
   A comprehensive SQL toolkit containing 20 Senior-level queries. This file is a masterclass in:

Advanced Cleaning: Handling NULLs using Window Functions (AVG() OVER) and Z-Score outlier detection.

Tech Synergy: Analyzing which combinations of Programming Languages and DL Frameworks yield the highest success.

Statistical Reporting: Executive-level summaries and cumulative performance tracking.