Benchmark Accuracy

Compute exact-match accuracy of model outputs: the fraction of positions where the predicted string equals the expected string after removing leading and trailing whitespace. Pairs are formed with zip, so extra items in the longer list are ignored.

Implement solve(predictions,targets).

Returns. Return a float in $[0,1][0,1]$.

Examples

Example 1

Input:

solve(['cat', 'dog', 'yes'], ['cat', 'cat', 'yes'])

Output:

0.666667

Example 2

Input:

solve([' cat ', 'Dog'], ['cat', 'dog'])

Output:

0.5