---

## 📂 Dataset Description

The dataset simulates anonymized customer-level credit card data and is split into training and testing files.

### `train.csv`

This file contains the features for each customer, along with the churn label (`Attrition_Flag`).

**Columns include:**

* `CLIENTNUM`: Unique customer identifier
* `Customer_Age`: Age of the customer
* `Gender`: Gender of the customer
* `Dependent_count`: Number of dependents
* `Education_Level`: Education level of the customer
* `Marital_Status`: Marital status
* `Income_Category`: Income bracket
* `Card_Category`: Type of credit card
* `Months_on_book`: How long the account has been active
* `Total_Relationship_Count`: Number of products customer is engaged with
* `Months_Inactive_12_mon`: Number of inactive months in the past year
* `Contacts_Count_12_mon`: Number of contacts with the bank in the past year
* `Credit_Limit`: Credit limit of the account
* `Total_Revolving_Bal`: Customer’s revolving balance
* `Avg_Open_To_Buy`: Average open-to-buy amount
* `Total_Amt_Chng_Q4_Q1`: Change in transaction amount from Q1 to Q4
* `Total_Trans_Amt`: Total transaction amount
* `Total_Trans_Ct`: Total number of transactions
* `Total_Ct_Chng_Q4_Q1`: Change in transaction count from Q1 to Q4
* `Avg_Utilization_Ratio`: Average card utilization ratio
* `loan_defaulted`: Target variable — 1 if the customer churned, 0 otherwise

**Sample:**

```csv
CLIENTNUM,Customer_Age,Gender,Dependent_count,loan_defaulted
768805383,45,M,3,0
818770008,49,F,5,0
```

---

### `test.csv`

Contains the same features as `train.csv` except for the `loan_defaulted` column. Your job is to predict this target variable.
**Note:** For submission, you must use the `id` column in the final CSV to match the required format.


---

### `sample_submission.csv`

Use this format for submitting your predictions. Each row must include:
`id` : From the test set
`loan_defaulted` : Your predicted label (0 = not churned, 1 = churned)

**Sample:**

```csv
id,loan_defaulted
768805383,0
818770008,1
```

Make sure your submission file exactly matches this format, or it will be rejected.

---