📘
Early Stopping Check
EasyDeep Learning
Problem Statement
Determine if training should stop based on validation loss history.
Background
Early stopping is a regularization technique that stops training when the model stops improving on a validation set. This prevents overfitting to the training data.
The rule: stop if validation loss hasn't improved for patience consecutive epochs.
Your Task
Write a function should_stop(history_losses, patience) that returns True if training should stop.
Output Format
Return a boolean: True if should stop, False otherwise.
Example:
Input:
history_losses=[5.0, 4.0, 4.1, 4.2], patience=2
Output:
True
Reasoning:
Best loss was 4.0 at index 1. Current index is 3. Epochs since best: 3-1=2. Since 2 >= patience(2), return True.
Constraints:
- len(history_losses) >= 1
- patience >= 1
- All losses are positive numbers
Editor
Python 3.13.1
Test Results
0/0Run code to see test results.