barplot.py

import pandas as pd
import matplotlib.pyplot as plt

# Load the dataset
df = pd.read_csv('Trips_by_Distance.csv')

def plot_participants_by_trip_distance():
    # Define the trip distance categories
    trip_categories = [
        "Number of Trips <1",
        "Number of Trips 1-3",
        "Number of Trips 3-5",
        "Number of Trips 5-10",
        "Number of Trips 10-25",
        "Number of Trips 25-50",
        "Number of Trips 50-100",
        "Number of Trips 100-250",
        "Number of Trips 250-500",
        "Number of Trips >=500",
    ]

    # Calculate the total number of participants in each category
    total_participants = df[trip_categories].sum()

    # Create a bar chart
    plt.figure(figsize=(12, 8))
    plt.bar(trip_categories, total_participants, color='skyblue', alpha=0.8)
    plt.xlabel("Trip Distance Category")
    plt.ylabel("Total Participants")
    plt.title("Total Participants by Trip Distance Category")
    plt.xticks(rotation=45, ha="right")
    plt.grid(True, axis='y')
    plt.tight_layout()
    plt.show()

# Invoke the plotting function
plot_participants_by_trip_distance()