Module 1 homework. ML Zoomcamp course.

mod_01_hw.py

"""Module 1 homework. ML Zoomcamp course."""

from pprint import pprint

import numpy as np
import pandas as pd


def main():
    df = get_df_from_url()
    results = {
        "q1": q1(),
        "q2": q2(df),
        "q3": q3(df),
        "q4": q4(df),
        "q5": q5(df),
        "q6": q6(df),
        "q7": q7(df),
    }

    pprint(results)


def get_df_from_url():
    url = "https://raw.githubusercontent.com/alexeygrigorev/datasets/master/laptops.csv"
    return pd.read_csv(url)


def q1():
    """Pandas version

    What's the version of Pandas that you installed?
    """
    return pd.__version__


def q2(df):
    """Records count
    How many records are in the dataset?
    """
    return df.shape[0]


def q3(df):
    """Laptop brands
    How many laptop brands are presented in the dataset?
    """

    return df["Brand"].nunique()


def q4(df):
    """Missing values
    How many columns in the dataset have missing values?
    """
    return (df.isna().sum() > 0).sum()


def q5(df):
    """Maximum final price

    What's the maximum final price of Dell notebooks in the dataset?
    """

    return df.query("Brand=='Dell'")["Final Price"].max()


def q6(df):
    """Maximum final price
    What's the maximum final price of Dell notebooks in the dataset?
    """
    median = df["Screen"].median()
    mode = df["Screen"].mode().values[0]
    median_fillna = df["Screen"].fillna(mode).median()
    has_it_changed = not (median == mode == median_fillna)
    return has_it_changed


def q7(df):
    """Sum of weights
    - Select all the "Innjoo" laptops from the dataset.
    - Select only columns RAM, Storage, Screen.
    - Get the underlying NumPy array. Let's call it X.
    - Compute matrix-matrix multiplication between the transpose of X and X. To get the transpose, use X.T. Let's call the result XTX.
    - Compute the inverse of XTX.
    - Create an array y with values [1100, 1300, 800, 900, 1000, 1100].
    - Multiply the inverse of XTX with the transpose of X, and then multiply the result by y. Call the result w.
    - What's the sum of all the elements of the result?
    - Note: You just implemented linear regression. We'll talk about it in the next lesson.
    """

    X = df.query("Brand=='Innjoo'")[["RAM", "Storage", "Screen"]].values
    XTX = X.T.dot(X)
    inverse_XTX = np.linalg.inv(XTX)
    y = np.array([1100, 1300, 800, 900, 1000, 1100])
    w = inverse_XTX.dot(X.T).dot(y)
    return w.sum()


if __name__ == "__main__":
    main()