In this challenge, you will use **polars** to efficiently perform data aggregation on large datasets. Specifically, you are tasked with grouping a large DataFrame by one column and computing the mean of another column for each group. This is a common operation in data analysis, especially when working with big data, as it allows you to summarize and extract insights from subsets of your data without loading everything into memory at once.

import unittest
import sys
import subprocess
import importlib
import ast
import re

# Ensure polars is installed before importing
try:
    import polars as pl
except ImportError:
    subprocess.check_call([sys.executable, "-m", "pip", "install", "polars"])
    import polars as pl

class TestTask(unittest.TestCase):
    def setUp(self):
        self.module = importlib.import_module("user_code")

    def test_returns_dataframe_with_one_row_per_group(self):
        df = pl.DataFrame({
            "group": ["X", "Y", "X", "Z", "Y", "Z", "X"],
            "score": [5, 10, 15, 20, 30, 40, 25]
        })
        result = self.module.groupby_mean(df, "group", "score")
        _dynamic_test(
            self,
            isinstance(result, pl.DataFrame),
            "Returned object is a polars DataFrame",
            f"Returned object is not a polars DataFrame: {type(result)}"
        )
        unique_groups = set(df["group"].to_list())
        _dynamic_test(
            self,
            result.height == len(unique_groups),
            "Returned DataFrame has one row per unique group value",
            f"Expected {len(unique_groups)} groups, got {result.height} rows"
        )

    def test_correct_mean_computation(self):
        df = pl.DataFrame({
            "team": ["A", "B", "A", "B", "A"],
            "points": [10, 20, 30, 40, 50]
        })
        result = self.module.groupby_mean(df, "team", "points")
        expected = {
            "A": (10 + 30 + 50) / 3,
            "B": (20 + 40) / 2
        }
        for row in result.iter_rows(named=True):
            group = row["team"]
            mean_val = row.get("mean_points")
            _dynamic_test(
                self,
                group in expected and abs(mean_val - expected[group]) < 1e-6,
                f"Mean for group {group} is correct",
                f"Expected mean {expected[group]} for group {group}, got {mean_val}"
            )

    def test_column_naming(self):
        df = pl.DataFrame({
            "cat": ["a", "b", "a"],
            "val": [1, 2, 3]
        })
        result = self.module.groupby_mean(df, "cat", "val")
        _dynamic_test(
            self,
            "mean_val" in result.columns,
            "Returned DataFrame column is named correctly as 'mean_val'",
            f"Returned DataFrame columns: {result.columns}"
        )

    def test_works_with_any_dataframe(self):
        df = pl.DataFrame({
            "city": ["NY", "LA", "NY", "SF"],
            "temp": [70, 80, 60, 65],
            "humidity": [30, 40, 50, 60]
        })
        # Test grouping by city, mean of temp
        result = self.module.groupby_mean(df, "city", "temp")
        expected = {
            "NY": (70 + 60) / 2,
            "LA": 80,
            "SF": 65
        }
        for row in result.iter_rows(named=True):
            group = row["city"]
            mean_val = row.get("mean_temp")
            _dynamic_test(
                self,
                group in expected and abs(mean_val - expected[group]) < 1e-6,
                f"Mean for city {group} is correct",
                f"Expected mean {expected[group]} for city {group}, got {mean_val}"
            )

def _dynamic_test(test_case, condition, success_message, failure_message):
    if condition:
        test_case._testMethodName = success_message
        test_case.assertTrue(True, success_message)
    else:
        test_case._testMethodName = failure_message
        test_case.fail(failure_message)

def normalize_text(text):
    text = text.lower()
    text = re.sub(r"\\s{2,}", " ", text)
    text = re.sub(r"\\s*([,:?])\\s*", r"\\1 ", text)
    return text.strip()

def change_var(code: str, var_name: str, value: str) -> str:
    tree = ast.parse(code)
    lines = code.splitlines()
    changed = False
    # Collect all assignment nodes to modify
    assign_nodes = [
        (i, node)
        for i, node in enumerate(tree.body)
        if isinstance(node, ast.Assign)
        and any(isinstance(target, ast.Name) and target.id == var_name for target in node.targets)
    ]

    # If nothing to change, return unmodified code
    if not assign_nodes:
        return code

    # Perform replacements for all matching assignments (from last to first to not break line offsets)
    for i, node in reversed(assign_nodes):
        start_line = node.lineno - 1
        line = lines[start_line]
        indent = ' ' * (len(line) - len(line.lstrip()))
        lines[start_line] = f"{indent}{var_name} = {value}"
        next_line = len(lines)
        for next_node in tree.body[i+1:]:
            if hasattr(next_node, 'lineno'):
                next_line = next_node.lineno - 1
                break
        if next_line > start_line + 1:
            lines[start_line+1:next_line] = []
        changed = True

    return '\\n'.join(lines) if changed else code

if __name__ == "__main__":
    unittest.main()


test_main.py

A practical, hands-on course for aspiring data scientists ready to tackle real-world large data challenges. Learn to efficiently process, sample, and analyze massive datasets using Python and essential libraries. Each section features engaging video explanations and interactive challenges to build your expertise.

Learn foundational strategies for handling datasets too large to fit in memory, including chunking and streaming techniques.

Explore methods to balance and sample large datasets, including oversampling and undersampling.

Learn to use the polars library for fast, memory-efficient data processing.

Challenge: Polars Data Aggregation

Lösung