Choose the input for the next decoder step. With teacher forcing (use_target=True) feed the ground-truth target; otherwise feed the model's own predicted output.
Implement solve(target, predicted, use_target).
Returns. Return target if use_target is true, else predicted (the chosen object itself, not a copy).
Example 1
Input:
solve([1, 2, 3], [0, 1, 2], True)
Output:
[1, 2, 3]
Example 2
Input:
solve([1, 2, 3], [0, 1, 2], False)
Output:
[0, 1, 2]
No results yet
Run your code to check it against the examples, or submit to run every test.