2024 AI Tools - Step-by-Step Guide

Amazon SageMaker

Amazon SageMaker is a fully managed service introduced by Amazon Web Services (AWS) in November 2017. Its purpose is to make it easier for developers and data scientists to build, train, and deploy machine learning (ML) models at scale. Before SageMaker, creating and deploying ML models required significant expertise and manual effort in setting up infrastructure, managing data, and optimizing models.

Create a New SageMaker Project:

Access SageMaker Studio: Once logged in, from the AWS Management Console, search for SageMaker in the search bar and click on SageMaker Studio.
Set Up SageMaker Studio: If it’s your first time using SageMaker Studio, you’ll need to create a SageMaker Domain by selecting a user profile, instance type, and permissions.
After the setup, click on Launch Studio to open the SageMaker Studio interface.

Create a New Notebook:

Start a Notebook: In the SageMaker Studio dashboard, click on File > New > Notebook. Choose the environment (like a Python 3 kernel) for the notebook.
Configure Notebook Settings: Specify instance type, networking, and storage if needed.
Initialize the Notebook: Click Create Notebook to start a new SageMaker notebook where you can write and execute machine learning code.

Train a Model:

Import Data: Use Amazon S3 to store and access the dataset you want to use for training.
Select an Algorithm: You can either use a built-in algorithm or import your own custom algorithm.
Start Training: Define the training job by specifying the algorithm, dataset, instance type, and hyperparameters. Run the training job directly from your notebook.

Amazon SageMaker Console:
https://aws.amazon.com/sagemaker/

Amazon SageMaker Studio Overview:
https://docs.aws.amazon.com/sagemaker/latest/dg/studio.html

Creating a SageMaker Domain (First-time setup for SageMaker Studio):
https://docs.aws.amazon.com/sagemaker/latest/dg/onboard-quick-start.html

Getting Started with Amazon SageMaker Notebooks:
https://docs.aws.amazon.com/sagemaker/latest/dg/notebooks.html

Training a Model in SageMaker:
https://docs.aws.amazon.com/sagemaker/latest/dg/how-it-works-training.html

Microsoft Copilot Studio

Step-by-Step Guide for Microsoft Copilot Studio

Open your browser and navigate to the Microsoft Copilot Studio URL https://copilotstudio.microsoft.com/. Log in using your Microsoft account credentials (company or organization).

Create a New Project: On the home screen, click on the «Create» button located on the left sidebar. You will see various templates such as «Safe Travels,» «Store Operations,» «Sustainability Insights,» «Voice,» «Weather,» and «Team Navigator.» Choose a template that best fits your project’s needs. For NLU, «Voice» might be a suitable choice if your project involves voice capabilities. «Team Navigator» is ideal if your project focuses on enhancing team collaboration and task management through AI-driven insights.«Store Operations» could be a great option if your project involves optimizing retail operations or inventory management.«Sustainability Insights» is suitable for projects aiming to track and analyze environmental data to support sustainability efforts.«Safe Travels» can be used for projects that involve travel safety and logistics, helping manage itineraries, safety protocols, or transportation data.«Weather» is perfect if your project requires weather-related predictions and data, useful for industries like agriculture or logistics.

Define Project Details: Provide a name, icon, language, etc. This helps in identifying and managing the project effectively. Click on the «Create» button to initialize the project.

Configure Data Sources: Once the project is created, you may be prompted to connect data sources. Depending on the project, this could involve linking databases, external APIs, or sample datasets provided within the template.

Train and Test the Model:

If the template requires machine learning models, such as for Voice or Team Navigator, you will need to train the model with your data.
Test the model using inbuilt test environments to ensure the configurations work as expected.

Deploy the Project:

Once your model is tested and ready, deploy the project by navigating to the Deploy section.
Specify any final settings such as compute resources and user access before confirming the deployment.
You will then receive a URL endpoint or integration instructions for using the project in a live environment.

Additional Resources for Microsoft Copilot Studio:

Microsoft Copilot Studio:
https://copilotstudio.microsoft.com/

Getting Started with Microsoft Copilot:
Microsoft Copilot Overview

(This link covers general information about Microsoft Copilot tools, but specific documentation for Copilot Studio may be available within the platform itself.)

Google Vertex AI

Step-by-Step Guide for Google Vertex AI

1. Open your browser and navigate to the Google Cloud Console:
https://console.cloud.google.com/vertex-ai.
Log in using your Google Cloud account credentials.

2. Enable Vertex AI API:

Once logged in, if this is your first time using Vertex AI, you may need to enable the Vertex AI API.
Go to the API & Services dashboard and search for Vertex AI. Enable it for your project.

3. Create a New Project:

In the Google Cloud Console, click the Project Selector (next to the Google Cloud logo at the top).
Click on New Project, provide a project name, organization (if needed), and a billing account. Click Create to initialize your project.

4. Access Vertex AI:

After creating the project, in the left-hand menu, click on AI & Machine Learning and select Vertex AI from the dropdown.

5. Create a New Vertex AI Model:

In the Vertex AI dashboard, click Create New to start building a model.
Select AutoML or Custom Training based on your needs:

– AutoML: For beginners, choose AutoML to let Google automatically train and optimize your model.
– Custom Training: If you have your own custom models or datasets, you can choose this option for more advanced use cases.

6. Import Your Data:

Select Dataset from the Vertex AI menu. Click on Create Dataset.
Choose your data type (e.g., Tabular, Image, Text, or Video) and upload your dataset from your local machine or use data stored in Google Cloud Storage.

7. Train Your Model:

After your data is uploaded, click on Train New Model. Choose a training method (AutoML or custom model training).
Define the target (what you’re trying to predict), configure the training options, and start the training process.

8. Deploy Your Model:

Once the model is trained, go to the Models section in the Vertex AI dashboard.
Click Deploy and specify the compute resources for serving your model (e.g., instance type and region).
After deployment, Vertex AI will provide you with an endpoint for making predictions.

9. Make Predictions:

After deployment, you can test your model by making real-time predictions from the Prediction tab.
Alternatively, you can use the provided REST API endpoint to integrate Vertex AI predictions into your application.

Additional Resources for Google Vertex AI:

Google Vertex AI Documentation:
https://cloud.google.com/vertex-ai/docs

Google Cloud Console (Vertex AI):
https://console.cloud.google.com/vertex-ai

Getting Started with AutoML:
https://cloud.google.com/vertex-ai/docs/start/automl-users

Custom Model Training:
https://cloud.google.com/vertex-ai/docs/training

Azure AI Studio

Azure AI Studio – Step-by-Step Guide

Open your browser and navigate to the Azure AI Studio URL https://portal.azure.com/#create/Microsoft.AI.

Create a New AI Service:

Once logged in, from the Azure Portal, click on Create a resource.
Search for AI + Machine Learning in the search bar, and select Azure AI from the results.
Click on the Create button to start configuring your new AI resource.

Select the AI Service Type:

Azure AI Studio offers various services such as Language Understanding (LUIS), Form Recognizer, Text Analytics, Speech, and Translator.
For Natural Language Processing (NLP), choose Language Understanding (LUIS) if your project involves analyzing text or intent recognition.

Define Project Details:

Provide a name, resource group, region, and other necessary information.
Select a pricing tier that matches your needs, then click “Review + Create” to finalize and deploy the AI resource.

Connect Data Sources:

Once the AI service is deployed, you can upload or link your data for training and testing the model.
In Form Recognizer, for example, you can upload PDFs or images to extract structured data.

Test and Deploy:

After configuring the data sources, test the AI model using the Test environment in Azure AI Studio.
When satisfied with the results, you can deploy the service by clicking on Deploy and specify the compute resources required for the deployment.

Azure Machine Learning Studio

Step-by-Step Guide

Open your browser and navigate to the Azure Machine Learning Studio URL https://ml.azure.com/.

Create a New Machine Learning Workspace:

On the Azure ML Studio dashboard, click Create a new workspace.
Fill in the workspace name, select a subscription, resource group, and region.
Click Create to initialize your workspace.

Upload a Dataset:

After creating the workspace, go to the Datasets section on the left sidebar.
Click Create dataset and choose whether to upload from a local file, cloud storage (Azure Blob Storage), or other sources.
Define the dataset details such as name, description, and data type, and then complete the upload.