Score prediction using Machine Learning algorithm?

Have you ever noticed how cricket analyst predict the score of a cricket match before it is played.So, this project is all about using ML algorithm to predict first innings score based on the inputs.

So let's straight "drive" into code:

Importing essential libraries

import pandas as pd
import pickle

Data file

df = pd.read_csv('ipl.csv')

Remove unwanted columns

columns_to_remove = ['mid', 'venue', 'batsman', 'bowler', 'striker', 'non-striker']
df.drop(labels=columns_to_remove, axis=1, inplace=True)

Keeping only consistent teams ,the teams who regularly play in ipl

consistent_teams = ['Kolkata Knight Riders', 'Chennai Super Kings', 'Rajasthan Royals',
                    'Mumbai Indians', 'Kings XI Punjab', 'Royal Challengers Bangalore',
                    'Delhi Daredevils', 'Sunrisers Hyderabad']
df = df[(df['bat_team'].isin(consistent_teams)) & (df['bowl_team'].isin(consistent_teams))]

Removing the first 5 overs data in every match , our prediction will be on overs more than 5

df = df[df['overs']>=5.0]

Converting date column to date time object

from datetime import datetime
df['date'] = df['date'].apply(lambda x: datetime.strptime(x, '%Y-%m-%d'))

Converting into dummies

encoded_df = pd.get_dummies(data=df, columns=['bat_team', 'bowl_team'])

Rearranging the columns

encoded_df = encoded_df[['date', 'bat_team_Chennai Super Kings', 'bat_team_Delhi Daredevils', 'bat_team_Kings XI Punjab',
              'bat_team_Kolkata Knight Riders', 'bat_team_Mumbai Indians', 'bat_team_Rajasthan Royals',
              'bat_team_Royal Challengers Bangalore', 'bat_team_Sunrisers Hyderabad',
              'bowl_team_Chennai Super Kings', 'bowl_team_Delhi Daredevils', 'bowl_team_Kings XI Punjab',
              'bowl_team_Kolkata Knight Riders', 'bowl_team_Mumbai Indians', 'bowl_team_Rajasthan Royals',
              'bowl_team_Royal Challengers Bangalore', 'bowl_team_Sunrisers Hyderabad',
              'overs', 'runs', 'wickets', 'runs_last_5', 'wickets_last_5', 'total']]

Splitting the data into train and test set

X_train = encoded_df.drop(labels='total', axis=1)[encoded_df['date'].dt.year <= 2016]
X_test = encoded_df.drop(labels='total', axis=1)[encoded_df['date'].dt.year >= 2017]

y_train = encoded_df[encoded_df['date'].dt.year <= 2016]['total'].values
y_test = encoded_df[encoded_df['date'].dt.year >= 2017]['total'].values

Removing the date column

X_train.drop(labels='date', axis=True, inplace=True)
X_test.drop(labels='date', axis=True, inplace=True)

Linear Regression Model

from sklearn.linear_model import LinearRegression
regressor = LinearRegression()
regressor.fit(X_train,y_train)

Creating a pickle file for our model

filename = 'first-innings-score-lr-model.pkl'
pickle.dump(regressor, open(filename, 'wb'))

Importing essential libraries

from flask import Flask, render_template, request
import pickle
import numpy as np

Loading our model

filename = 'first-innings-score-lr-model.pkl'
regressor = pickle.load(open(filename, 'rb'))

app = Flask(__name__)

@app.route('/')
def home():
    return render_template('index.html')

@app.route('/predict', methods=['POST'])
def predict():
    temp_array = list()

    if request.method == 'POST':

        batting_team = request.form['batting-team']
        if batting_team == 'Chennai Super Kings':
            temp_array = temp_array + [1,0,0,0,0,0,0,0]
        elif batting_team == 'Delhi Daredevils':
            temp_array = temp_array + [0,1,0,0,0,0,0,0]
        elif batting_team == 'Kings XI Punjab':
            temp_array = temp_array + [0,0,1,0,0,0,0,0]
        elif batting_team == 'Kolkata Knight Riders':
            temp_array = temp_array + [0,0,0,1,0,0,0,0]
        elif batting_team == 'Mumbai Indians':
            temp_array = temp_array + [0,0,0,0,1,0,0,0]
        elif batting_team == 'Rajasthan Royals':
            temp_array = temp_array + [0,0,0,0,0,1,0,0]
        elif batting_team == 'Royal Challengers Bangalore':
            temp_array = temp_array + [0,0,0,0,0,0,1,0]
        elif batting_team == 'Sunrisers Hyderabad':
            temp_array = temp_array + [0,0,0,0,0,0,0,1]

        bowling_team = request.form['bowling-team']
        if bowling_team == 'Chennai Super Kings':
            temp_array = temp_array + [1,0,0,0,0,0,0,0]
        elif bowling_team == 'Delhi Daredevils':
            temp_array = temp_array + [0,1,0,0,0,0,0,0]
        elif bowling_team == 'Kings XI Punjab':
            temp_array = temp_array + [0,0,1,0,0,0,0,0]
        elif bowling_team == 'Kolkata Knight Riders':
            temp_array = temp_array + [0,0,0,1,0,0,0,0]
        elif bowling_team == 'Mumbai Indians':
            temp_array = temp_array + [0,0,0,0,1,0,0,0]
        elif bowling_team == 'Rajasthan Royals':
            temp_array = temp_array + [0,0,0,0,0,1,0,0]
        elif bowling_team == 'Royal Challengers Bangalore':
            temp_array = temp_array + [0,0,0,0,0,0,1,0]
        elif bowling_team == 'Sunrisers Hyderabad':
            temp_array = temp_array + [0,0,0,0,0,0,0,1]

        overs = float(request.form['overs'])
        runs = int(request.form['runs'])
        wickets = int(request.form['wickets'])
        runs_in_prev_5 = int(request.form['runs_in_prev_5'])
        wickets_in_prev_5 = int(request.form['wickets_in_prev_5'])

        temp_array = temp_array + [overs, runs, wickets, runs_in_prev_5, wickets_in_prev_5]

        data = np.array([temp_array])
        my_prediction = int(regressor.predict(data)[0])

        return render_template('result.html', lower_limit = my_prediction-10, upper_limit = my_prediction+5)

if __name__ == '__main__':
    app.run(port=5500)