Analyze a year of your spending

Goal

Load a CSV bank statement, group by month + category with pandas, and plot the trends with matplotlib. Deliver a `spending.png` chart and a printed summary showing your top three leak categories.

Why it matters

pandas is what makes Python worth learning — and the best way to learn it is on your own data. You'll look at this chart, then change your habits.

Requirements · rubric

Build against this exact list — it is the spec your published project is judged on.

• 01Load a CSV (any Indian bank export or a sample you generate)
• 02Auto-detect date column format; parse into datetime
• 03Categorize transactions (rules-based or keyword-based — no ML required)
• 04Aggregate spend by (month, category) using pandas groupby
• 05Plot a stacked-bar or line chart to `spending.png`
• 06Print top 3 categories by total spend, with percentages
• 07Handle at least 1 year of data (12+ months) gracefully

Starter code

Copy this into a new file — or open it in a workspace as analyze.py. The TODO parts are yours to fill.

"""
Spending analyzer.

Usage:
    python analyze.py statement.csv

Outputs:
    spending.png — stacked bar chart of monthly category totals
    Prints top 3 leak categories to stdout.
"""

from __future__ import annotations
import sys
from dataclasses import dataclass
from pathlib import Path

import pandas as pd
import matplotlib.pyplot as plt


CATEGORIES = {
    "food":     ["swiggy", "zomato", "restaurant", "cafe", "starbucks"],
    "transport": ["uber", "ola", "petrol", "diesel", "irctc", "metro"],
    "shopping":  ["amazon", "flipkart", "myntra", "ajio"],
    "bills":     ["airtel", "jio", "electricity", "gas", "internet"],
    "rent":      ["rent", "landlord"],
}


@dataclass
class Summary:
    top_categories: list[tuple[str, float]]
    total_spend: float


def load(csv_path: Path) -> pd.DataFrame:
    # TODO: read_csv, parse dates, coerce numeric amount column
    # statement.csv (seeded next to this file) has: date, description, amount
    raise NotImplementedError("step 1, load(): read the CSV into a DataFrame")


def categorize(df: pd.DataFrame) -> pd.DataFrame:
    # TODO: add 'category' column based on CATEGORIES keyword rules
    # Fall back to "other" if nothing matches.
    raise NotImplementedError("step 2, categorize(): add a category column")


def aggregate(df: pd.DataFrame) -> pd.DataFrame:
    # TODO: groupby([month, category]) -> sum(amount). Return a wide pivot.
    raise NotImplementedError("step 3, aggregate(): monthly totals per category")


def plot(pivot: pd.DataFrame, out: Path) -> None:
    # TODO: stacked bar chart, save to out.
    raise NotImplementedError("step 4, plot(): save a stacked bar chart")


def summarize(df: pd.DataFrame) -> Summary:
    # TODO: return Summary with top 3 categories + total spend
    raise NotImplementedError("step 5, summarize(): top 3 categories and total")


def main() -> None:
    # In the workspace there is no command line, so default to the sample
    # statement seeded next to this file. Locally: python analyze.py mine.csv
    csv_path = Path(sys.argv[1]) if len(sys.argv) > 1 and sys.argv[1] else Path("statement.csv")
    df = load(csv_path)
    df = categorize(df)
    pivot = aggregate(df)
    plot(pivot, Path("spending.png"))
    s = summarize(df)
    print(f"Total spend: ₹{s.total_spend:,.0f}")
    for cat, amt in s.top_categories:
        pct = amt / s.total_spend * 100
        print(f"  {cat:12s}  ₹{amt:>10,.0f}  ({pct:.1f}%)")


if __name__ == "__main__":
    try:
        main()
    except NotImplementedError as todo:
        # A fresh starter is SUPPOSED to stop here. Say which step is next
        # instead of printing a traceback that looks like a bug.
        print(f"Not built yet: {todo}")
        print("Write that function, then press Run again. Each step you finish moves this message forward.")