## 📊 **Evaluation Criteria – Digital Dash**

How your model performs matters—but how it's **measured** is just as important. Here's how we’ll evaluate your submissions during the competition.

---

### 🧪 **Evaluation Metric**
Submissions will be evaluated using the following metric:

> **[Insert Metric Here — e.g., Accuracy, RMSE, F1 Score, AUC-ROC, Log Loss, MAE, etc.]**

Make sure your model is optimized for this metric, as it will directly determine your position on the leaderboard.

---

### 🧩 **Leaderboard Splits**
- The test set is divided into two parts:
  - **Public Leaderboard (X%)** – used to display live rankings during the competition.
  - **Private Leaderboard (Y%)** – used for final rankings after the competition ends.
- Your **best public submission** will be re-evaluated on the **private test set** to determine final winners.

_Example: 40% public / 60% private split._

---

### 📤 **Submission Format**
Your submission should be a file in the format:

```csv
id, prediction
001, 0.85  
002, 0.13  
...  
```

- Ensure the column names and order match the sample submission file.
- Any mismatch in format may result in submission failure or errors.

---

### 🔁 **Submission Limits**
- **Max 5 submissions per day.**
- Only your **best-scoring public submission** will be used for final private leaderboard evaluation.

---

### ⚠️ **Tie-breaker Rules**
In case of a tie on the private leaderboard:
1. The **earliest submission** with the top score will be ranked higher.
2. If needed, **code clarity, documentation, and reproducibility** will be reviewed to determine the winner.