Week 2 - Deploying a Face Emotion Detector on AWS Lambda

🛍️ Overview

Humans have always had the innate ability to recognize and distinguish between faces. Now computers are able to do the same. This opens up tons of applications. Face detection and Recognition can be used to improve access and security, allow payments to be processed without physical cards, enable criminal identification and allow personalized healthcare and other services. Face detection and recognition is a heavily researched topic and there are tons of resources online.

In today's session, we'll be taking a look at face detection and emotion interpretation. Our goal will be to detect the faces in an image and identify the emotions those faces portray. To do this, we'll need some tools!

Facial detection is the first part of our pipeline. We have used the Open-CV python library for Face Recognition using the Haar Cascade object detection algorithm that we found easy to install and very accurate in detecting faces. This library scans the input image and returns the bounding box coordinates of all detected faces. You will work on writing the code to implement this algorithm in emotions.py.

We then will utilize a pre-trained model and create the associated layers in a python class. This will allow us to utilize the model to make our predictions. You will work on writing the code to implement the layers for this class in emotions.py

We will then prepare SAM CLI to work with AWS Lambda. You will build your application using SAM's build function. This will utilize Docker to prepare your application to be uploaded to the Elastic Container Registry (ECR). The ECR is a repository for Docker or other containers, which we will discuss more further in Week 4.

After, we'll navigate to ECR on AWS to create our own private registry. We will then use SAM's deploy function to push our image to ECR and deploy our function on AWS Lambda. The resulting endpoint will be used to make an inference using Postman about our input image.

📚 Learning Objectives

By the end of this session, you will be able to:

Implement and utilize pre-trained models for your own project
Utilize Docker to build an Image of your application and upload to ECR
Use the SAM CLI tool to work with AWS Lambda
Deploy your application on AWS Lambda to make predictions

📦 Deliverables

Make the project public on your GitHub
Submit your repo link on Canvas
Submit a image prediction of your deployed model

AWS Lambda

What is AWS Lambda?

AWS Lambda is a serverless infrastructure and computing services. This means you do not need to acquire a server or any other infrastructure. The coder simply writes the code, uploads it, and runs it. It is part of core compute services (EC2, EBS, ELB, Lambda) proved by AWS. Amazon manages the computer power, but it is serverless from our perspective. AWS handles all the underlying infrastructure, deployment of code, and scaling.

What languages are supported by AWS Lambda?

python, Java, node.js, and C#

Example usage of AWS Lambda

Whenever you upload files, you want to create thumbnails automatically. AWS Lambda can do this for you. Every time you upload files, a Lambda function is called and thumbnail is generated. Other uses include:

scheduled tasks
microservices
event handlers
replace simple applications
custom near real-time process of logs
respond to events generated by S3

Billing for AWS Lambda

Pay for compute time per 100ms. Never pay for idle resources.

How do you use AWS Lambda?

1.) Create functions through inline editor or upload zip file.

Start a function
Define runtime settings
Define memory limit
Define execution time out

2.) Invoke those functions through CLI or SDK, or events

3.) The code is usually event-driven

The Program

This project contains source code and supporting files for a serverless application that you can deploy with the SAM CLI. It includes the following files and folders.

app.py - Code for the application's Lambda function and Project Dockerfile.
events - Invocation events that you can use to invoke the function.
tests - Unit tests for the application code.
template.yaml - A template that defines the application's AWS resources.

The application uses several AWS resources, including Lambda functions and an API Gateway API. These resources are defined in the template.yaml file in this project. You can update the template to add AWS resources through the same deployment process that updates your application code.

Deploying the Sample Application

The Serverless Application Model Command Line Interface (SAM CLI) is an extension of the AWS CLI that adds functionality for building and testing Lambda applications. It uses Docker to run your functions in an Amazon Linux environment that matches Lambda. It can also emulate your application's build environment and API.

To use the SAM CLI, you need the following tools.

SAM CLI - Install the SAM CLI
Docker - Install Docker community edition

You may need the following for local testing.

Python 3 installed

To build and deploy your application for the first time, run the following in your shell:

sam build
sam deploy --guided

The first command will build a docker image from a Dockerfile and then copy the source of your application inside the Docker image. The second command will package and deploy your application to AWS, with a series of prompts:

Stack Name: The name of the stack to deploy to CloudFormation. This should be unique to your account and region, and a good starting point would be something matching your project name.
AWS Region: The AWS region you want to deploy your app to.
Confirm changes before deploy: If set to yes, any change sets will be shown to you before execution for manual review. If set to no, the AWS SAM CLI will automatically deploy application changes.
Allow SAM CLI IAM role creation: Many AWS SAM templates, including this example, create AWS IAM roles required for the AWS Lambda function(s) included to access AWS services. By default, these are scoped down to minimum required permissions. To deploy an AWS CloudFormation stack which creates or modifies IAM roles, the CAPABILITY_IAM value for capabilities must be provided. If permission isn't provided through this prompt, to deploy this example you must explicitly pass --capabilities CAPABILITY_IAM to the sam deploy command.
Save arguments to samconfig.toml: If set to yes, your choices will be saved to a configuration file inside the project, so that in the future you can just re-run sam deploy without parameters to deploy changes to your application.

You can find your API Gateway Endpoint URL in the output values displayed after deployment.

Use the SAM CLI to Build and Test Locally

Build your application with the sam build command.

facebot$ sam build

The SAM CLI builds a docker image from a Dockerfile and then installs dependencies defined in requirements.txt inside the docker image. The processed template file is saved in the .aws-sam/build folder.

Test a single function by invoking it directly with a test event. An event is a JSON document that represents the input that the function receives from the event source. Test events are included in the events folder in this project.

Run functions locally and invoke them with the sam local invoke command.

facebot$ sam local invoke FaceSentimentFunction --event events/event.json

The SAM CLI can also emulate your application's API. Use the sam local start-api to run the API locally on port 3000.

facebot$ sam local start-api
facebot$ curl http://localhost:3000/

The SAM CLI reads the application template to determine the API's routes and the functions that they invoke. The Events property on each function's definition includes the route and method for each path.

      Events:
        Root:
          Type: Api
          Properties:
            Path: /
            Method: ANY

Add a Resource to Your Application

The application template uses AWS Serverless Application Model (AWS SAM) to define application resources. AWS SAM is an extension of AWS CloudFormation with a simpler syntax for configuring common serverless application resources such as functions, triggers, and APIs. For resources not included in the SAM specification, you can use standard AWS CloudFormation resource types.

Fetch, Tail, and Filter Lambda Function Logs

To simplify troubleshooting, SAM CLI has a command called sam logs. sam logs lets you fetch logs generated by your deployed Lambda function from the command line. In addition to printing the logs on the terminal, this command has several nifty features to help you quickly find the bug.

NOTE: This command works for all AWS Lambda functions; not just the ones you deploy using SAM.

facebot$ sam logs -n sentiment-analysis --stack-name facebot --tail

You can find more information and examples about filtering Lambda function logs in the SAM CLI Documentation.

Unit Tests

Tests are defined in the tests folder in this project. Use PIP to install the pytest and run unit tests from your local machine.

facebot$ pip install pytest pytest-mock --user
facebot$ python -m pytest tests/ -v

Cleanup

To delete the sample application that you created, use the AWS CLI. Assuming you used your project name for the stack name, you can run the following:

aws cloudformation delete-stack --stack-name facebot

Resources

See the AWS SAM developer guide for an introduction to SAM specification, the SAM CLI, and serverless application concepts.

Next, you can use AWS Serverless Application Repository to deploy ready to use Apps that go beyond hello world samples and learn how authors developed their applications: AWS Serverless Application Repository main page

yingjiema/MLOp4-W2