Design an **early stopping** check that decides when to halt training based on validation loss over epochs. You’ll implement a small utility that returns the first epoch where training should stop and the epoch with the best validation loss.

An epoch $t$ is considered an improvement if:
$$
L_t < \min_{i < t}(L_i) - \delta
$$

## Requirements

Implement the function:
```python
def early_stopping_epoch(val_losses, patience, min_delta):
    \"""
    Determines early stopping epoch using validation losses.

    Args:
        val_losses (np.ndarray): Validation loss at each epoch, in order.
        patience (int): Number of consecutive non-improving epochs allowed.
        min_delta (float): Minimum decrease in loss required to count as improvement.

    Returns:
        tuple[int, int]:
            (stop_epoch, best_epoch)
            - stop_epoch: the first epoch index where early stopping triggers, or -1 if it never triggers
            - best_epoch: the epoch index with the lowest validation loss (earliest in case of ties)
    \"""
    # Your solution below
```

**Rules:**
- Track the best (lowest) validation loss seen so far and its epoch index.
- An epoch is “improving” only if it beats the best loss by at least `min_delta`.
- Early stopping triggers when you see `patience` consecutive epochs without improvement.
- Return `stop_epoch` as the epoch where the patience counter first reaches `patience`.
- Use **NumPy** for input type, but basic Python iteration is sufficient.

## Example

```python
val_losses = np.array([1.00, 0.95, 0.96, 0.97, 0.94, 0.95, 0.96])
patience = 2
min_delta = 0.0

early_stopping_epoch(val_losses, patience, min_delta)
```

Output:
```python
(3, 1)
```

Class Survival Counts

Mean Claimant Age by Gender

Early stopping criterion

Solve this easy level ML coding question from General. The premier LeetCode for ML platform to master Python, SQL, and Machine Learning algorithms.

1106. Early stopping criterion

Examples

Constraints

Similar Questions

Argument	Type
patience	int
min_delta	float
val_losses	np.ndarray