AWS Artificial Intelligence & Machine Learning (AIML) Workshop for .NET Developers

This repository contains sample code for the AWS .NET AIML workshop. You need an AWS account to try the labs. If you are a .NET developer interested in exploring how to use AWS Artificial Intelligence & Machine Learning (AIML) to your uses cases feel free to drop an email to dotnet-aiml at amazon.com. We can organize an event for your team and provide free AWS accounts to try the labs.

• Lab setup guide (prerequisite for all the labs)
• Amazon Bedrock for .NET Developers
• Amazon Polly for .NET Developers
• Amazon Fraud Detector for .NET Developers

How to get your AWS enviornment ready.

You can try the labs in your own account or using an AWS provided account. How to get your lab enviornment ready can be found here

How the projet is organized

1. In the command prompt, run the following command

mkdir C:\Dev
cd c:\Dev
git clone https://github.com/aws-samples/dotnet-aiml-samples.git
cd C:\Dev\dotnet-aiml-samples\samples
start .

2. Once the git repo is cloned, open the folder C:\Dev\dotnet-aiml-samples\Samples. Open the Visual Studio solution file Samples.sln using Visual Studio 2022.

3. All samples are just console applications you can run independently.

   • Samples.Common: A class library containing helpful utility functions.
   • Samples.Bedrock: Amazon Bedrock samples
   • Samples.Bedrock.Rag: A sample Retrieval-Augmented Generation (RAG) solution.
   • Samples.Polly: Code samples to generate voice from text.

4. In Samples.Common->Utility.cs

   public static AWSCredentials GetCredentials()
   {
       //Make sure you create a profile using AWS CLI and save access key & secrete key
       //watch https://www.youtube.com/watch?v=fwtmTMf53Ek for more information
       string profileName = "my-dev-profile";
       var chain = new CredentialProfileStoreChain();
       AWSCredentials awsCredentials = null;
       if (!chain.TryGetAWSCredentials(profileName, out awsCredentials))
       {
           Console.WriteLine($"No profile name {profileName}  is found. Using the default credentials");
           awsCredentials = FallbackCredentialsFactory.GetCredentials();
       }
   
       return awsCredentials;
   }

This checks the existence of a profile called my-dev-profile; if it exists, it will use the credentials provided in the profile. Otherwise, it will use the credentials provided by the EC2 instance profile. To learn how to create AWS access & secret keys and save them in a profile watch this video

4. In Samples.Common->ISample This interface is implemented by all the samples. You can run the sample by creating an instance and calling the ISample->Run command.

   //An interface to test run a sample code fragment
   public interface ISample
   {
       //Run the sample code
       public void Run();
   }

For example, if you open Samples.Bedrock->Program.cs or Samples.Polly->Program.cs you will notice a code fragment like below

//An interface to test run a sample code fragment
static void Main(string[] args)
{
   AWSCredentials creds = Utility.GetCredentials();
   ISample s1 = new Sample1(creds);
   ISample s2 = new Sample2(creds);
   ...
   
   s1.Run();
   //s2.Run();
   ...
}

This initiate an instance of the sample S1 & run it. When you want to try Sample 2, comment the s1.Run() and comment out s2.Run(). Then, save the file, compile, and run the project.

License

This library is licensed under the MIT-0 License. See the LICENSE file.