""" Chapter-05-Matplotlib-Basics - Matplotlib: Line, Scatter, Histogram, Box Plot =============================================================================== File : Session5-Matplotlib-V01.py Version : V01 Date : 2026-05-13 Objective --------- Load a people CSV from the shared A-Data folder, create four classic Matplotlib chart types (line, scatter, histogram, box plot), and save each chart as a 300-DPI PNG file to the C-Results/Session5 output folder. What Is Matplotlib? -------------------- Matplotlib is Python's most widely used plotting library. It follows a "figure -> axes -> elements" model: plt.figure() Create a new blank canvas (figure). plt.plot() etc. Draw chart elements onto the current figure. plt.title/xlabel/... Add labels and formatting. plt.tight_layout() Auto-adjust spacing so labels do not overlap. plt.savefig(path) Write the figure to a PNG (or PDF, SVG, ...) file. plt.close() Release the figure from memory. Always call plt.close() after saving so memory is freed; otherwise each figure stays open and can accumulate into a memory leak in long scripts. Four Chart Types Introduced in This Session -------------------------------------------- Chart type Function Best used for --------------- ------------------- ------------------------------------ Line plot plt.plot() Trends over a continuous axis (time, sorted numeric variable like Age). Scatter plot plt.scatter() Relationship / correlation between two numeric variables. Histogram plt.hist() Distribution shape of one numeric variable; how often each range occurs. Box plot plt.boxplot() Spread, median, and outliers of one or more numeric variables side-by-side. Key Parameters Explained -------------------------- Parameter Used in Meaning ---------------------- --------------- --------------------------------- marker='o' plt.plot() Draw a circle at each data point. grid(True, lw=0.3) all Show faint background grid lines. bins=5 plt.hist() Number of equal-width buckets. edgecolor='black' plt.hist() Draw a border around each bar. axis='y' plt.grid() Grid lines on y-axis only. labels=[...] plt.boxplot() Labels under each box. dpi=300 savefig() High-resolution output (300 dots per inch; suitable for print). Box Plot Anatomy ----------------- ---- <- upper whisker (up to 1.5 * IQR above Q3) +--+ | | <- box spans Q1 (25%) to Q3 (75%) = interquartile range (IQR) +--+ <- median line inside the box (50%) | | +--+ ---- <- lower whisker (down to 1.5 * IQR below Q1) o <- individual outlier point (beyond whiskers) Expected Folder Layout ---------------------- (root)/ | +-- A-Data/ | +-- People.csv <- Input data (read-only) | +-- Chapter-05-Matplotlib-Basics/ | +-- Session5-Matplotlib-V01.py <- THIS script | +-- C-Results/ +-- Session5/ +-- line_income_vs_age.png <- Line chart output +-- scatter_height_weight.png <- Scatter chart output +-- hist_income.png <- Histogram output +-- box_height_weight.png <- Box plot output Required Input File (People.csv) ----------------------------------- The CSV must contain at least these four numeric columns: Age Age in years. Income Annual income in USD. Height_cm Height in centimetres. Weight_kg Weight in kilograms. Sample file content (People.csv): Name,Age,Income,Height_cm,Weight_kg Ali,25,45000,175,70 Sara,30,62000,160,55 John,22,38000,180,75 Mei,28,55000,165,60 Luis,35,71000,170,68 Quick Start ----------- 1. Open VS Code and open the Chapter-05-Matplotlib-Basics folder. 2. Open Session5-Matplotlib-V01.py. 3. Run (Ctrl+F5, or right-click -> Run Python File in Terminal). 4. Four PNG files appear in C-Results/Session5/. 5. Open each PNG to inspect the charts. Step-by-Step Flow ----------------- Step 1 : Reconfigure stdout/stderr to UTF-8 for cross-platform printing. Step 2 : Import required libraries: sys, os, pathlib.Path, pandas, matplotlib. Step 3 : Resolve the input CSV path (two levels up, then into A-Data/). Step 4 : Resolve the project root and build the Session5 output folder path. Step 5 : Create the output folder if it does not already exist. Step 6 : Print resolved paths for verification. Step 7 : Load the CSV into a pandas DataFrame. Step 8 : Sort the DataFrame by Age for the line plot. Step 9 : Create and save Chart 1 -- Line plot: Income vs Age. Step 10 : Create and save Chart 2 -- Scatter plot: Height vs Weight. Step 11 : Create and save Chart 3 -- Histogram: Income distribution. Step 12 : Create and save Chart 4 -- Box plot: Height and Weight. Step 13 : Print a success message listing all four saved PNG file paths. Output Files ------------- line_income_vs_age.png Line chart showing Income on Y-axis vs Age on X-axis, sorted by Age. Markers at each data point. 300 DPI. scatter_height_weight.png Scatter plot of Height (X) vs Weight (Y). Each dot is one person. Good for spotting a positive correlation between the two variables. hist_income.png Histogram of Income divided into 5 equal-width bins. Shows whether incomes cluster in a range or are spread out. box_height_weight.png Side-by-side box plots of Height_cm and Weight_kg. Shows median, interquartile range, and any outliers. Notes ----- - .values on a pandas Series converts it to a plain NumPy array before passing to plt.plot(). This avoids a pandas index mismatch warning when the DataFrame has been sorted (the index is reordered but not reset). - plt.close() must be called after each savefig() to free figure memory. Skipping it causes figures to accumulate -- harmless for 4 charts but a real problem in loops that create hundreds of figures. - dpi=300 produces print-quality images. Use dpi=72 or dpi=96 for smaller files suited to web or screen display. - grid(True, linewidth=0.3) uses a very thin line so the grid is visible but does not compete visually with the data. - os.path.join() is used for the savefig() paths here because savefig() also accepts a pathlib.Path directly -- either style works. """ # =========================================================================== # Step 1 - Reconfigure stdout and stderr to UTF-8 # Prevents garbled or missing characters on Windows terminals. # The hasattr() guard keeps this safe on older Python builds. # =========================================================================== import sys if hasattr(sys.stdout, 'reconfigure'): sys.stdout.reconfigure(encoding='utf-8') if hasattr(sys.stderr, 'reconfigure'): sys.stderr.reconfigure(encoding='utf-8') # =========================================================================== # Step 2 - Import required libraries # =========================================================================== import os # Used for os.path.join() in savefig paths from pathlib import Path # Modern object-oriented path handling import pandas as pd # DataFrame for data loading import matplotlib.pyplot as plt # Core Matplotlib plotting interface # =========================================================================== # Step 3 - Resolve the input CSV path # # Path(__file__).resolve() -> absolute path of THIS script file # .parents[1] -> two levels up (project root) # / 'A-Data' / 'People.csv' -> navigate into the shared data folder # # This means the CSV is located at: # (root)/A-Data/People.csv # regardless of which directory the terminal is pointed at when you run. # =========================================================================== data_path = Path(__file__).resolve().parents[1] / 'A-Data' / 'People.csv' print(f'[DEBUG] data_path : {data_path}') # =========================================================================== # Step 4 - Resolve the project root and build the output folder path # # A dedicated Session5 subfolder keeps this session's PNG files separate # from outputs produced by other sessions. # =========================================================================== project_root = Path(__file__).resolve().parents[1] # (root)/ out_folder = project_root / 'C-Results' / 'Session5' # (root)/C-Results/Session5/ print(f'[DEBUG] project_root : {project_root}') print(f'[DEBUG] out_folder : {out_folder}') # =========================================================================== # Step 5 - Create the output folder if it does not already exist # parents=True -> also create C-Results/ if it is missing. # exist_ok=True -> do nothing silently if the folder is already there. # =========================================================================== out_folder.mkdir(parents=True, exist_ok=True) # =========================================================================== # Step 6 - Verify the input file exists before trying to load it # Giving a clear message here is more helpful than letting pandas # raise a raw FileNotFoundError with a long traceback. # =========================================================================== if not data_path.exists(): print(f'[ERROR] Input file not found: {data_path}') print('[ERROR] Check that People.csv is in the A-Data folder. Exiting.') raise SystemExit(1) print(f'[INFO] Input file found: {data_path}') # =========================================================================== # Step 7 - Load the CSV into a pandas DataFrame # encoding='utf-8' handles accented or special characters in names. # =========================================================================== print('\n[INFO] Loading data from CSV...') df = pd.read_csv(data_path, encoding='utf-8') print(f'[DEBUG] Shape loaded : {df.shape[0]} rows x {df.shape[1]} cols') print(f'[DEBUG] Columns found : {list(df.columns)}') print(df.head(), '\n') # =========================================================================== # Step 8 - Sort the DataFrame by Age for the line plot # # df.sort_values('Age') # Returns a new DataFrame with rows ordered from youngest to oldest. # Without sorting, plt.plot() connects the dots in CSV row order, which # produces a jagged zigzag line rather than a smooth left-to-right trend. # # .values on a sorted Series converts it to a plain NumPy array. # This is important because sort_values reorders the pandas index # (e.g. [2, 0, 4, 1, 3]) without resetting it. If we passed the Series # directly, Matplotlib would receive mismatched indices and print a warning. # Using .values strips the index entirely, giving Matplotlib a clean array. # =========================================================================== df_sorted = df.sort_values('Age') print(f'[DEBUG] DataFrame sorted by Age.') # =========================================================================== # Step 9 - Chart 1: Line plot -- Income vs Age # # plt.figure() # Creates a new blank figure canvas. Every chart starts with this call # to ensure each chart is independent and does not draw on a previous one. # # plt.plot(x, y, marker='o') # Draws a line connecting the data points in order. # marker='o' adds a filled circle at each data point so individual # values are visible as well as the connecting line. # # plt.grid(True, linewidth=0.3) # Draws faint horizontal and vertical grid lines to help read values. # linewidth=0.3 keeps them subtle so they do not overpower the data. # # plt.tight_layout() # Automatically adjusts margins so axis labels, tick labels, and the # title are not clipped or overlapping. Always call this before savefig. # # plt.savefig(path, dpi=300) # Writes the figure to a PNG file at 300 dots per inch (print quality). # # plt.close() # Releases the figure from memory. Must be called after every savefig() # to prevent memory accumulation when generating many charts. # =========================================================================== print('[INFO] Creating Chart 1: Line plot - Income vs Age...') plt.figure() plt.plot( df_sorted['Age'].values, # X-axis: Age values as plain NumPy array df_sorted['Income'].values, # Y-axis: Income values as plain NumPy array marker='o' # Circle marker at each data point ) plt.title('Income vs Age (sorted by Age)') plt.xlabel('Age') plt.ylabel('Income (USD)') plt.grid(True, linewidth=0.3) plt.tight_layout() out_line = out_folder / 'line_income_vs_age.png' plt.savefig(out_line, dpi=300) plt.close() # Free figure memory after saving print(f'[DEBUG] Saved: {out_line}') # =========================================================================== # Step 10 - Chart 2: Scatter plot -- Height vs Weight # # plt.scatter(x, y) # Draws one dot per row at position (Height_cm, Weight_kg). # Unlike plt.plot(), scatter does NOT connect the dots with lines, # making it ideal for showing the relationship between two variables # without implying any ordering. # # A cluster of dots trending from lower-left to upper-right indicates a # positive correlation: taller people tend to weigh more. # =========================================================================== print('[INFO] Creating Chart 2: Scatter plot - Height vs Weight...') plt.figure() plt.scatter( df['Height_cm'], # X-axis: Height in centimetres df['Weight_kg'] # Y-axis: Weight in kilograms ) plt.title('Height vs Weight') plt.xlabel('Height (cm)') plt.ylabel('Weight (kg)') plt.grid(True, linewidth=0.3) plt.tight_layout() out_scatter = out_folder / 'scatter_height_weight.png' plt.savefig(out_scatter, dpi=300) plt.close() print(f'[DEBUG] Saved: {out_scatter}') # =========================================================================== # Step 11 - Chart 3: Histogram -- Income distribution # # plt.hist(series, bins=5, edgecolor='black') # Divides the range of Income into 5 equal-width buckets (bins) and # counts how many values fall into each bucket. # edgecolor='black' draws a visible border around each bar, making # adjacent bars easier to distinguish. # # plt.grid(axis='y', linewidth=0.3) # Grid lines on the Y-axis only (horizontal lines) so you can read # the count values. Vertical grid lines are suppressed because the # bars already span the X-axis range. # # A histogram answers: "Are incomes clustered around a typical value, # or spread across a wide range?" # =========================================================================== print('[INFO] Creating Chart 3: Histogram - Income distribution...') plt.figure() plt.hist( df['Income'], # One bar per bin covering the Income range bins=5, # Divide the range into 5 equal buckets edgecolor='black' # Black border around each bar ) plt.title('Income Distribution') plt.xlabel('Income (USD)') plt.ylabel('Count') plt.grid(axis='y', linewidth=0.3) # Horizontal grid lines only plt.tight_layout() out_hist = out_folder / 'hist_income.png' plt.savefig(out_hist, dpi=300) plt.close() print(f'[DEBUG] Saved: {out_hist}') # =========================================================================== # Step 12 - Chart 4: Box plot -- Height and Weight side-by-side # # plt.boxplot([list_of_series], labels=[list_of_names]) # Draws one box for each Series in the list. # Passing a Python list of Series (not a DataFrame) lets us place two # columns with very different numeric ranges (cm vs kg) on the same # figure for a visual comparison of their spread. # # Each box shows five summary statistics: # - Bottom whisker : minimum non-outlier value # - Bottom of box : Q1 (25th percentile) # - Line in box : median (50th percentile) # - Top of box : Q3 (75th percentile) # - Top whisker : maximum non-outlier value # - Circles (o) : individual outlier points beyond 1.5 * IQR # # labels=['Height_cm', 'Weight_kg'] places a readable label under each box. # =========================================================================== print('[INFO] Creating Chart 4: Box plot - Height and Weight...') plt.figure() plt.boxplot( [df['Height_cm'], df['Weight_kg']], # Two boxes: one per column labels=['Height_cm', 'Weight_kg'] # Labels displayed below each box ) plt.title('Box Plots: Height and Weight') plt.tight_layout() out_box = out_folder / 'box_height_weight.png' plt.savefig(out_box, dpi=300) plt.close() print(f'[DEBUG] Saved: {out_box}') # =========================================================================== # Step 13 - Print a success message listing all four saved PNG file paths # =========================================================================== print('\n[DONE] All chart files saved:') print(f' 1. {out_line}') print(f' 2. {out_scatter}') print(f' 3. {out_hist}') print(f' 4. {out_box}')