Create a linear regression model that predicts the outcome for a tennis player based on their playing habits. By analyzing and modeling the Association of Tennis Professionals (ATP) data, you will determine what it takes to be one of the best tennis players in the world.

tennis_ace_challenge.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:
players = pd.read_csv('tennis_stats.csv')
print(players.head(5))

# perform exploratory analysis here:
plt.scatter(players.BreakPointsOpportunities, players.Wins)
plt.title('Break Points Opportunities and Wins')
plt.show()
plt.clf()

plt.scatter(players.BreakPointsOpportunities, players.Losses)
plt.title('Break Points Opportunities and Losses')
plt.show()
plt.clf()

plt.scatter(players.Aces, players.Wins)
plt.title('Aces and Wins')
plt.show()
plt.clf()

# helpers
def write_model(features, outcome, title):
  features = players[features]
  outcome = players[outcome]
  features_train, features_test, outcome_train, outcome_test = train_test_split(features, outcome, train_size=0.8)
  model = LinearRegression()
  model.fit(features_train, outcome_train)
  print(model.score(features_test, outcome_test))
  prediction = model.predict(features_test)
  plt.scatter(outcome_test, prediction, alpha=0.4)
  plt.title(title)
  plt.xlabel('Actual')
  plt.ylabel('Predicted')
  plt.legend()
  plt.show()
  plt.clf()

## perform single feature linear regressions here:
write_model(['BreakPointsOpportunities'], 'Winnings', 'Break Points Opportunities to Winnings')

#Aces
write_model(['Aces'], 'Winnings', 'Aces to Winnings')

## perform two feature linear regressions here:
#Aces
write_model(['BreakPointsOpportunities', 'Aces'], 'Winnings', 'Break Points Opportunities and Aces to Winnings')

# BP Ops & Ranking
write_model(['BreakPointsOpportunities', 'FirstServeReturnPointsWon'], 'Ranking', 'Break Points Opportunities and 1st Serve Return Points to Winnings')

# FirstServeReturnPointsWon
write_model(['BreakPointsOpportunities', 'FirstServeReturnPointsWon'], 'Winnings', 'Break Points Opportunities and 1st Serve Return Points to Winnings')

## perform multiple feature linear regressions here:
write_model(['FirstServe','FirstServePointsWon','FirstServeReturnPointsWon',
'SecondServePointsWon','SecondServeReturnPointsWon','Aces',
'BreakPointsConverted','BreakPointsFaced','BreakPointsOpportunities',
'BreakPointsSaved','DoubleFaults','ReturnGamesPlayed','ReturnGamesWon',
'ReturnPointsWon','ServiceGamesPlayed','ServiceGamesWon','TotalPointsWon',
'TotalServicePointsWon'], ['Winnings'], 'All Features')

write_model(['FirstServe','FirstServePointsWon','FirstServeReturnPointsWon',
'SecondServePointsWon','SecondServeReturnPointsWon','Aces',
'BreakPointsConverted','BreakPointsFaced','BreakPointsOpportunities',
'BreakPointsSaved','DoubleFaults','ReturnGamesPlayed','ReturnGamesWon',
'ReturnPointsWon','ServiceGamesPlayed','ServiceGamesWon','TotalPointsWon',
'TotalServicePointsWon'], ['Ranking'], 'All Features')