Lazy Bob has created a machine learning model which receives one number as the input and outputs that number + 1 (a very interesting model, as you can see).
He has a list of $$$N$$$ ($$$1 ≤ N ≤ 10,000$$$) expected outputs $$$e_i$$$ and actual outputs $$$a_i$$$ ($$$1 ≤ e_i, a_i ≤ 10^9$$$), and he wants to find out the accuracy of his model. If an actual output is within $$$K$$$ ($$$0 ≤ K ≤ 100,000$$$) of its corresponding expected output, he counts it as the same (his model is not that good).
Output the model's accuracy as a percentage to the nearest integer (for example, if 8 out of 12 cases work as expected, that would be considered an accuracy of 67%) but do NOT include the percent sign in your answer. Check the Notes section if this is unclear.
Line 1: Two space-separated integers, $$$N$$$ and $$$K$$$.
Line 2…$$$N+1$$$: 2 space-separated integers $$$e_i$$$ and $$$a_i$$$.
Line 1: A percentage representing Bob's model's accuracy (to the nearest integer). It is guaranteed that the accuracy will never be < 0.01 away from the middle of two integers.
5 3 3 6 7 3 1 10 8 7 11 11
60
$$$1 ≤ N ≤ 10,000$$$
$$$0 ≤ K ≤ 100,000$$$
$$$1 ≤ a_i, e_i ≤ 10^9$$$
3 out of the 5 output pairs are within K of each other, which is why the accuracy is ⅗*100% = 60%. Drop the percent sign to get an answer of 60.
