Uber Rides Prediction using machine learning

As ride-hailing services become more popular, being able to reliably estimate demand can help operators allocate drivers to clients more efficiently, reducing idle time, improving traffic congestion, and improving the passenger experience.

In this Blog, we are making a Flask based web application that will predict the Number of Weekly Rides using machine learning model.

For making this Appliation we mainly divide the project in two steps

1. Training Model
2. Deploy the trained model In the Flask app

Before discussed more make sure you have the following prerequisites installed in your personal machine (Computer/Laptop).

Flask
numpy
scikit-learn
pandas, pickle

For Model Traning, we import some necessary libraries. Read the dataset using the pandas library. after reading the dataset we generally divide the dataset into training and test split by using sklearn train_test_split.

Then trained a most popular machine learning model name Linear regression. Basically for uber ride prediction, the target labels Number of weekly riders are continuous values. So we need a regression-based model to predict the unknown outcomes.

After training the linear regression model we save the trained model in a pickle file. The code for the training model is given below.

Training Model Code

import pandas as pd
# import numpy as np
import pickle
from sklearn.model_selection import train_test_split
from sklearn.linear_model import LinearRegression

data = pd.read_csv('uber_dataset.csv')
# print(data.head())

data_x = data.iloc[:,0:-1].values
data_y = data.iloc[:,-1].values
print(data_y)

X_train,X_test,y_train,y_test = train_test_split(data_x,data_y,test_size=0.3,random_state=0)

reg = LinearRegression()
reg.fit(X_train,y_train)

print("Train Score:", reg.score(X_train,y_train))
print("Test Score:", reg.score(X_test,y_test))

pickle.dump(reg, open('model.pkl','wb'))

model = pickle.load(open('model.pkl','rb'))
print(model.predict([[80, 1770000, 6000, 85]]))

After training a model. It’s time to make the complete user-friendly web application to test our trained model. First, we import some necessary libraries then load the model using pickle again. We make a prediction function that takes input from the user and predicts the output

The code for testing model uisng flask app is shown below.

Flask code

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

app = Flask(__name__)
model = pickle.load(open('model.pkl','rb'))

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

@app.route('/predict', methods=['POST'])
def predict():
    int_features  = [int(x) for x in request.form.values()]
    final_features = [np.array(int_features)]
    prediction = model.predict(final_features)
    output = round(prediction[0],2)
    return render_template('index.html',prediction_text="Number of Weekly Rides Should be {}".format(math.floor(output)))

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

Outputs

Complete Project Download Link: