Deployment of a Machine Learning model into Amazon EC2(Part 2- Real-Time Deployment)

by Pronay Ghosh and Hiren Rupchandani

• In the previous article, we set up a Linux-based EC2 instance so that we can deploy our model on AWS EC2.
• Let’s proceed with model deployment!

Creating a Security Group:

• Now that we have our EC2 instance, we will have to set up the EC2 instance accessible from anywhere.

Security Group

• To do so, go to the EC2 instance and select the security group.
• Select Inbound and click on add rule.
• Set the type of the rule to All traffic.

Setting Inbound rules

• In the source information, set the permission as Anywhere.
• This indicates that we can use access the EC2 instance from anywhere with IPv4 and IPv6 network protocols.

Setting up the Network Interfaces:

• After setting up the security groups, we need to go to the Network Interfaces section.

Network Interfaces

• Right-click on the interface and select Change security groups.

Change Security groups

Setting up the system dependencies:

• Now, we need to go back to the terminal and install the system dependencies in order to deploy our model.
• For that, we will use the following command:

pip3 install -r requirements.txt

Installing using requirements.txt

• After running the command we can see that the system dependencies have been successfully installed.
• Upon successful installation, run the following command in the terminal:

python3 app.py

Running the app

• We can now see that our app is running on port number 5000

Deploying into EC2:

• As the final step, we will have to go back to the EC2 instances in the AWS management console.

Connecting the EC2 instance

• From the instances, select the EC2 instance.
• After that, we will have to click on actions and click on connect. We should be able to see an URL.

Getting the URL

• We will simply add the port number. Here, the port number is 5000. So we can add the port number as follows:

<The EC2 URL>/5000

• Open this entire link in a new tab and you can see your model live on Amazon Elastic Compute Cloud (EC2).

Conclusion:

• In this article, we covered a high-level overview of how to deploy a machine learning model into Amazon EC2.
• In the next article, we will see how the high-level overview of Amazon S3.
• We will also understand how to run a jupyter notebook into the cloud using Amazon Sagemaker.

Follow us for more upcoming future articles related to Data Science, Machine Learning, and Artificial Intelligence.

Also, Do give us a Clap👏 if you find this article useful as your encouragement catalyzes inspiration for and helps to create more cool stuff like this.