Deploy a Machine Learning application on AWS EC2 server !

6 min readJan 16, 2022

In this tutorial, we will create an ML model and pickle the model. Build a basic flask application and load the pickled model for the prediction. Deploy the app on local first to test the end to end run. Further deploy the application on AWS EC2 server.

Build an ML model and save the pickle file.
Build a flask application and load the pickled file for the prediction.
Deploy the flask application locally and test the model prediction.
Provision an AWS EC2 instance.
Deploy the flask application on EC2 instance and test the application.

Let’s start.

1. Build an ML model and save the pickle file.

Let’s build a simple Regression model to predict the salary of a candidate based on his/her experience, test score and interview score. Training data looks like this.

Let’s build the model and save the model as a pickle file.

import numpy as np
import matplotlib.pyplot as plt
import pandas as pd
import pickledataset = pd.read_csv('hiring.csv')dataset['experience'].fillna(0, inplace=True)dataset['test_score'].fillna(dataset['test_score'].mean(), inplace=True)X = dataset.iloc[:, :3]#Converting words to integer values
def convert_to_int(word):
    word_dict = {'one':1, 'two':2, 'three':3, 'four':4, 'five':5, 'six':6, 'seven':7, 'eight':8,
                'nine':9, 'ten':10, 'eleven':11, 'twelve':12, 'zero':0, 0: 0}
    return word_dict[word]X['experience'] = X['experience'].apply(lambda x : convert_to_int(x))y = dataset.iloc[:, -1]#Splitting Training and Test Set
#Since we have a very small dataset, we will train our model with all available data.from sklearn.linear_model import LinearRegression
regressor = LinearRegression()#Fitting model with trainig data
regressor.fit(X, y)# Saving model to disk
pickle.dump(regressor, open('model.pkl','wb'))

Let’s load a and predict the model.

model = pickle.load(open('model.pkl','rb'))
print(model.predict([[2, 9, 6]]))[53290.89255945]

2. Build a flask application and load the pickled file for the prediction.

Let’s build a sample flask application.

import numpy as np
from flask import Flask, request, jsonify, render_template
import pickleapp = 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():
    '''
    For rendering results on HTML GUI
    '''
    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='Employee Salary should be $ {}'.format(output))if __name__ == "__main__":
    app.run(debug=True)

index.html as a GUI template.

<!DOCTYPE html>
<html >
<!--From https://codepen.io/frytyler/pen/EGdtg-->
<head>
  <meta charset="UTF-8">
  <title>ML API</title>
  <link href='https://fonts.googleapis.com/css?family=Pacifico' rel='stylesheet' type='text/css'>
<link href='https://fonts.googleapis.com/css?family=Arimo' rel='stylesheet' type='text/css'>
<link href='https://fonts.googleapis.com/css?family=Hind:300' rel='stylesheet' type='text/css'>
<link href='https://fonts.googleapis.com/css?family=Open+Sans+Condensed:300' rel='stylesheet' type='text/css'>
  
</head><body>
 <div class="login">
 <h1>Predict Salary Analysis</h1><!-- Main Input For Receiving Query to our ML -->
    <form action="{{ url_for('predict')}}"method="post">
     <input type="text" name="experience" placeholder="Experience" required="required" />
        <input type="text" name="test_score" placeholder="Test Score" required="required" />
  <input type="text" name="interview_score" placeholder="Interview Score" required="required" /><button type="submit" class="btn btn-primary btn-block btn-large">Predict</button>
    </form><br>
   <br>
   {{ prediction_text }}</div></body>
</html>

3. Deploy the flask application locally and test the model prediction.

> flask run
 * Environment: production
   WARNING: This is a development server. Do not use it in a production deployment.
   Use a production WSGI server instead.
 * Debug mode: off
 * Running on http://127.0.0.1:5000/ (Press CTRL+C to quit)

Salary prediction on the UI.

4. Provision an AWS EC2 instance.

Please follow this guide to provision the EC2 instance on AWS cloud.

Get started with AWS cloud and launch your first EC2 instance ! | by Shekhar Khandelwal | Jan, 2022 | Medium

5. Deploy the flask application on EC2 instance and test the application.

First, lets provide required access for external applications to connect to the ec2 instance. Click on “Security Groups” under “Network & Security” in the aws console.

Click on “Create security group”.

Provide a name to the security group, and provide inbound and outbound rules. For this sample, let all traffic interact with the server.

Click on Create Security Group.

Now, click on “Network Intefaces”.

Right click on the network interface and select “Change security groups”.

Select the newly create “Full access” security group.

Click on “Add security group” and click on Save.

Let’s ssh into the EC2 instance and update/install required packages, if required.

[ec2-user@ip-172-31-44-84 ~]$ sudo yum update && yum install python3-pip

Create a folder flask-app in the server.

[ec2-user@ip-172-31-44-84 ~]$ pwd
/home/ec2-user
[ec2-user@ip-172-31-44-84 ~]$ mkdir flask-app
[ec2-user@ip-172-31-44-84 ~]$ ll
total 0
drwxrwxr-x 2 ec2-user ec2-user 6 Jan 16 13:22 flask-app
[ec2-user@ip-172-31-44-84 ~]$ cd flask-app
[ec2-user@ip-172-31-44-84 flask-app]$

Update the flask app.py file to make it transfer the files to the ec2 server.

Remove these lines.

if __name__ == "__main__":
    app.run(debug=True)

Add these lines.

if __name__ == "__main__":
    app.run(host="0.0.0.0", port="8080")

Basically, we are making the app to run on the local server and listen on port 8080 from the external applications.

Now copy app.py, model.pkl, requirements.txt and templates/index.html files on the ec2 instance.

> scp -i "..\EC2\sample-ec2-instance.pem" -r .\app.py .\model.pkl .\requirements.txt .\templates\ ec2-user@13.59.123.239:/home/ec2-
user/flask-app

ssh into the server and check the successful file transfer.

> ssh -i "..\EC2\sample-ec2-instance.pem" ec2-user@13.59.123.239

Run pip3 install command to install all dependencies.

pip3 install requirements.txt

Now, run flask app to deploy the flask application.

python3 app.py

Get the EC2 server hostname from the aws console.

And finally, use port 8080 to launch the application deployed on the server.

Congratulations, now, your application is running on the EC2 server. Let’s predict.

Happy Learning !

Github — AWS_series/Deploy_MLmodel_on_EC2 at main · shekharkhandelwal1983/AWS_series (github.com)

References — Tutorial 4- Deployment Of ML Models In AWS EC2 Instance — YouTube