from collections import namedtuple
You should put blank lines after imports, and between function/class definitions.
a = namedtuple('assignment', 'weight scores')
Using a namedtuple for a simple data class like this is a good idea, but it should still be named as a class; Assignment.
def average(scores:list):
'''average a list of scores'''
return sum(scores)/2
This function has an obvious bug; what about lists of scores with lengths other than two? Also you type the parameter but not the return value, and should again have some more whitespace; I'd write:
def average(scores: list) -> float:
"""Average a list of scores."""
return sum(scores) / len(scores)
See PEP-257 for docstring conventions.
def get_grade(grades:list):
'''compute final grade based on scores and their weights'''
result = 0
for a in grades:
result += average(a.scores) * a.weight
return result
I think you could simplify this with a bit more OOP. If we subclass the named tuple we can add a read only calculated attribute to it:
class Assignment(namedtuple('Assignment', 'weight scores')):
@property
def result(self) -> float:
return average(self.scores) * self.weight
Then your function becomes:
def get_grade(assignments: list) -> float:
"""Compute final grade based on scores and their weights."""
return sum(a.result for a in assignments)
Note the revised parameter name, which better describes the input.
def main():
grades = [a(.30, [18/45, 11/55]),
a(.20, [3/10, 7.5/10, 10/10]),
a(.10, [9/10, 9/10, 10/10, 10/10, 8/10,
6/10, 10/10, 10/10]),
a(.40, [29/65]),
a(.01, [1/1])]
print(get_grade(grades))
main()
To keep things looking consistent without introducing a lot of indentation, I usually split lists with long elements across multiple lines as follows:
assignments = [
Assignment(.30, [18/45, 11/55]),
...
Assignment(
.10,
[9/10, 9/10, 10/10, 10/10, 8/10, 6/10, 10/10, 10/10]
),
...
]
Note splitting between, rather than within, the two parameters.
Also you should guard the invocation of the entry point as follows:
if __name__ == '__main__':
main()
This means you can import this functionality more easily for testing and reuse.
