# **Dataset Description**

This section provides a detailed description of the data files and the information they contain. Understanding the dataset is crucial for effective feature engineering and predictive modeling. Below, you'll find the answers to common questions and an explanation of each file and column included in the dataset.

## **Files**
1. **`train.csv`** - The training set, containing H3 hexagon IDs and their corresponding cost of living values.
   - **Columns:**
     - `hex_id`: The H3 hexagon index representing a geographic area.
     - `cost_of_living`: A value between 0 and 1 indicating the cost of living in that hexagon area.

3. **`mobility.parquet`** - A dataset containing device-level mobility data, provided in Parquet format. Use this data to derive features for your model.
   - **Columns:**
     - `device_id`: An anonymized identifier for each mobile device.
     - `lat`: Latitude of the recorded location.
     - `lon`: Longitude of the recorded location.
     - `timestamp`: Timestamp indicating when the data point was recorded.

4. **`test.csv`** - A sample submission file provided to guide you in formatting your final predictions. Use this as a reference to ensure your submission matches the required structure.
   - **Columns:**
     - `hex_id`: The H3 hexagon index corresponding to a geographic area.
     - `cost_of_living`: Your predicted value for the cost of living, ranging from 0 to 1.

## **Columns in `train.csv`**
- **`hex_id`**: An H3 hexagon index that corresponds to a geographic area. Each hexagon covers a portion of the map and serves as a spatial aggregation unit for your predictions.
- **`cost_of_living`**: A value between 0 and 1 representing the cost of living for the respective hexagon area.

## **Columns in `test.csv`**
- **`hex_id`**: An H3 hexagon index that corresponds to a geographic area.
- **`cost_of_living`**: This column is empty and should be filled with your predictions.

## **Columns in `mobility_data.parquet`**
- **`device_id`**: A unique anonymized identifier for each mobile device in the dataset.
- **`lat`**: The latitude where the device was observed.
- **`lon`**: The longitude where the device was observed.
- **`timestamp`**: A UNIX timestamp representing the precise time when the observation occurred.

## **Frequently Asked Questions**
1. **What files do I need?**
   - You will primarily work with `train.csv` to build your model and `test.csv` to generate predictions. Use `mobility_data.parquet` to extract features that could improve the predictive power of your model.
   
2. **What should I expect the data format to be?**
   - The data files are in CSV and Parquet formats. Each row in `train.csv` and `test.csv` corresponds to a geographic area, while `mobility_data.parquet` contains device-level observations. Predictions are submitted as a CSV file matching the format in `test.csv`.

3. **What am I predicting?**
   - You are predicting the `cost_of_living`, a value between 0 and 1, for each geographic hexagon (`hex_id`) in the `test.csv` file.

4. **What acronyms will I encounter?**
   - **H3**: A geospatial indexing system that divides the world into hexagonal cells, used for representing geographic areas.
   - **CSV**: Comma-separated values, the format for `train.csv` and `test.csv`.
   - **Parquet**: A columnar storage file format, used for `mobility_data.parquet`.

## **Submission Example**
Below is an example of the expected CSV format for your submission:

```csv
hex_id,cost_of_living
8a2a1072b59ffff,0.45
8a2a1072a7bffff,0.60
8a2a1072959ffff,0.33
8a2a1072b0fffff,0.75
8a2a107282bffff,0.25
```