---

### 📁 **Dataset Description**

The competition uses the **MNIST (Modified National Institute of Standards and Technology)** dataset—one of the most well-known datasets in the field of machine learning and computer vision.

It contains **70,000 grayscale images** of handwritten digits from **0 to 9**, each sized **28x28 pixels**.

#### 📦 Dataset Split
- **Training Set**: 60,000 images with labels (used to train your model).
- **Test Set**: 10,000 images without labels (used for evaluation).

#### 🖼️ Image Format
- Each image is a **28x28 pixel grayscale image**, flattened into a 784-length vector in CSV format or as `.png` images depending on the dataset version used.
- Pixel values range from **0 (black)** to **255 (white)**.

#### 🧾 File Format (CSV)
If provided in CSV format, the files are structured as follows:

##### ✅ `train.csv`
| label | pixel0 | pixel1 | pixel2 | ... | pixel783 |
|-------|--------|--------|--------|-----|----------|
| 5     | 0      | 0      | 12     | ... | 0        |

- **label**: The correct digit (0 to 9).
- **pixel0 to pixel783**: Grayscale values of each pixel.

##### ❓ `test.csv`
| pixel0 | pixel1 | pixel2 | ... | pixel783 |
|--------|--------|--------|-----|----------|
| 0      | 0      | 12     | ... | 0        |

- **No label column** in test data. Your task is to predict the missing labels and submit them in the `submission.csv`.

---