Step 3: Recognise چالش 3 از جلسهی 12، متوسط
لیست training جفتهای (عکس، برچسب) دارد. تابع recognise(picture) بنویس که با k-NN (k=3) برچسب عکس تازه را برگرداند: فاصله تا هر عکس آموزشی، sort، سه تای اول، رأیگیری. برای TEST چاپ کن I see a 1.
def flatten(picture):
return [1 if ch == "#" else 0 for row in picture for ch in row]
def distance(a, b):
return sum(1 for i in range(len(a)) if a[i] != b[i])
training = [
([".#####.", "#.....#", "#.....#", "#.....#", "#.....#", "#.....#", ".#####."], 0),
(["..###..", ".#...#.", "#.....#", "#.....#", "#.....#", ".#...#.", "..###.."], 0),
([".####..", "#....#.", "#....#.", "#....#.", "#....#.", "#....#.", ".####.."], 0),
(["...#...", "..##...", "...#...", "...#...", "...#...", "...#...", ".#####."], 1),
(["...#...", "...#...", "...#...", "...#...", "...#...", "...#...", "...#..."], 1),
(["..#....", ".##....", "..#....", "..#....", "..#....", "..#....", ".####.."], 1),
]
TEST = ["...#...", "..##...", "...#...", "...#...", "...#...", "...#...", "..###.."]
def recognise(picture):
pass
print("I see a", recognise(TEST))
راهنما و جواب این چالش · برگشت به درس جلسهی 12: پروژه: تشخیص رقم کوچک · چالش بعدی: Step 4: Accuracy
بیشتر بخوان: list comprehension در پایتون: از حلقه for تا یک خط کد