Data science: Monthly totals and growth

Single sales arrive in any order. Turn them into monthly totals in time order, with the growth from each month to the next.

What to build

A program that totals amounts per month and prints the percentage change.

Requirements

  • Print one line per month that has data, in time order: YYYY-MM TOTAL, with the total to two decimals.
  • From the second month on, add a space and the percentage change from the previous printed month, with its sign and one decimal place, followed by %.

Input

Line 1: n. Then n lines YYYY-MM-DD AMOUNT.

Output

One line per month.

Constraints

1 ≤ n ≤ 1000; amounts are over 0.

Example

Input:

5
2024-01-05 100
2024-01-20 50
2024-02-03 180
2024-03-15 90
2024-03-01 45

Output:

2024-01 150.00
2024-02 180.00 +20.0%
2024-03 135.00 -25.0%

Why it works

Grouping by the YYYY-MM prefix aggregates each month, sorting the prefixes puts months in time order, and each change divides the difference by the earlier month's total.

Examples

Example 1 · Input
5
2024-01-05 100
2024-01-20 50
2024-02-03 180
2024-03-15 90
2024-03-01 45
Expected output
2024-01 150.00
2024-02 180.00 +20.0%
2024-03 135.00 -25.0%
Example 2 · Input
2
2024-01-01 10
2023-12-31 10
Expected output
2023-12 10.00
2024-01 10.00 +0.0%

Prompts inside input() are ignored when checking, so input("n: ") is fine.

Hints

Hint 1

The month of a date YYYY-MM-DD is date[:7].

Hint 2

Sort the months; YYYY-MM strings sort in time order.

Hint 3

f"{change:+.1f}%" prints the sign.

Solution

Show one correct solution (try it yourself first)
n = int(input())
totals = {}
for _ in range(n):
    date, amount = input().split()
    totals[date[:7]] = totals.get(date[:7], 0) + float(amount)
prev = None
for month in sorted(totals):
    line = f"{month} {totals[month]:.2f}"
    if prev is not None:
        line += f" {(totals[month] - prev) / prev * 100:+.1f}%"
    print(line)
    prev = totals[month]

Lesson: Time series basics →