Codecademy export

script.py

import codecademylib3_seaborn
import pandas as pd
import matplotlib.pyplot as plt
from sklearn.model_selection import train_test_split
from sklearn.linear_model import LinearRegression

# load and investigate the data here:
data=pd.read_csv('tennis_stats.csv')
print(data.head())
print(data.dtypes)
print(data.shape)
print(data.columns)










# perform exploratory analysis here:
plt.figure(figsize=[15,15])
for i in range(data.shape[1]-2):
  columns=data.columns[2:]
  plt.subplot(6, 4, i+1)
  name=columns[i]
  plt.title(f'Winnings vs {name}')
  plt.scatter(data.Winnings, data.iloc[:,[i+2]], norm=True, alpha=0.5)
plt.subplots_adjust(wspace=1.3, hspace=1.3)
plt.show()

x_train, x_test, y_train, y_test=train_test_split(data['ServiceGamesPlayed'], data['Winnings'], test_size=0.2, random_state=12)

import numpy as np
model=LinearRegression()
model.fit(np.array(x_train).reshape(-1, 1), y_train)
print(model.score(np.array(x_test).reshape(-1, 1), y_test))

plt.clf()
plt.scatter(data.ServiceGamesPlayed, data.Winnings, norm=True, alpha=0.5)
plt.show()

print(x_test.shape)
print(y_test.shape)
## perform single feature linear regressions here:
prediction=model.predict(np.array(x_test).reshape(-1, 1))
plt.scatter(np.array(x_test).reshape(-1,1), prediction, alpha=0.4 )
plt.show()




















## perform two feature linear regressions here:

X_train, X_test, Y_train, Y_test=train_test_split(data['BreakPointsOpportunities'], data['Winnings'], test_size=0.2, random_state=12)
model.fit(np.array(X_train).reshape(-1, 1), Y_train)
print(model.score(np.array(X_test).reshape(-1, 1), Y_test))

XX_train, XX_test, YY_train, YY_test=train_test_split(data.BreakPointsFaced, data.Winnings, test_size=0.2, random_state=12)
model.fit(np.array(XX_train).reshape(-1,1), YY_train)
print(model.score(np.array(XX_test).reshape(-1,1), YY_test))




















## perform multiple feature linear regressions here:
f_train, f_test, t_train, t_test=train_test_split(data.drop(['Player', 'Year', 'Winnings'], axis=1), data['Winnings'], test_size=0.2)

model.fit(f_train, t_train)
print(model.score(f_test, t_test))