blakezara / matplotlip Goto Github PK

# Dependencies
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns

# Import Files And Merge

ride_data = pd.read_csv("raw_data/ride_data.csv")
city_data = pd.read_csv("raw_data/city_data.csv")

merged = pd.merge(city_data, ride_data, how='outer', on='city')
merged.head()

merged.columns=("City", "Driver Count", "Type", "Date", "Fare", "Ride ID")


merged.head()

#groupby
merged_data = merged.groupby(["City","Type","Driver Count"])

#average fare
new_data = merged_data["Fare"].mean()
new_data = pd.DataFrame(new_data)

#total rides
new_data["Total Rides"] = merged_data["Ride ID"].nunique()
new_data.reset_index(inplace=True)

#reorganize
new_data.rename(columns={"Fare":"Average Fare"},inplace=True)

new_data.head()

size = np.arange(0, 1000, 10)

figure= sns.lmplot(x='Total Rides', y='Average Fare', hue='Type', scatter_kws={"s": size, 'alpha':0.50,}, data=new_data, fit_reg=False)

plt.ylim(15, 45)
plt.xlim(0, 40)
plt.xlabel('Total Number of Rides Per City')
plt.ylabel('Average Fare')

plt.title('Pyber Rideshare Data')

plt.show()

#Percent of Total Fares by City Type
type_data = merged.groupby("Type")['Type', 'Fare', 'Ride ID', 'Driver Count']

fare = type_data.sum()["Fare"]                           
fare

labels = fare.index
explode = [.3 , 0, 0]

plt.pie(fare, startangle = 140, explode = explode, labels = labels, 
        autopct = "%1.1f%%", 
        shadow = True, 
        wedgeprops = {'linewidth': .2, 'edgecolor': 'black'})

plt.title("Percentage of Total Fares by City Type")
plt.show()

#Percentage of Total Rides by City Type

rides= type_data.count()["Ride ID"]
rides

labels = rides.index
explode = [.3 , 0, 0]

plt.pie(rides, startangle = 140, explode = explode, labels = labels, autopct = "%1.1f%%", shadow = True, wedgeprops = {'linewidth': .5, 'edgecolor': 'black'})

plt.title("Percentage of Total Rides by City Type")

plt.show()

#Percentage of Total Drivers by City Type

drivers= type_data.sum()["Driver Count"]
drivers

labels = rides.index
explode = [.3 , 0, 0]

plt.pie(drivers, startangle = 140, explode = explode, labels = labels, autopct = "%1.1f%%", shadow = True, wedgeprops = {'linewidth': .5, 'edgecolor': 'black'})

plt.title("Percentage of Total Drivers by City Type")
plt.show()