Getting started with AI Apps

Last updated January 10th, 2022.

Objective

OVHcloud provides a set of managed AI tools designed for building your machine learning projects.

This guide explains how to get started with OVHcloud AI Apps by deploying your first application through the Control Panel.

Requirements

• access to the OVHcloud Control Panel
• a Public Cloud project in your OVHcloud account

Instructions

Subscribe to AI Apps

Log in to your OVHcloud Control Panel and switch to Public Cloud in the top navigation bar. After selecting your Public Cloud project, click on AI Apps in the left-hand navigation bar under AI & Machine Learning.

Click on the Deploy an AI App button and accept terms and conditions if any.

Once clicked, you will be redirected to the creation process detailed below.

Step 1: Select the location

Select where your AI App will be hosted, meaning the physical location.

Step 2: Select the application to deploy

AI Apps allows a user to deploy applications from two sources:

• From an OVHcloud catalog with already built-in AI models and applications.
• From your own Docker container, giving you the full flexibility to deploy what you want.

In this tutorial, we will select English Sentiment Analysis from the HuggingFace company.

These AI models will work to detect the positive or negative feelings in English sentences.

Step 3: Assign compute resources

An AI App needs compute resources to be deployed, from a minimum of one instance to a maximum of ten. Each instance will be billed based on the amount of running time, whether or not you use your AI App.

Each node is composed of:

• CPU or GPU resources
• Local storage
• Network bandwidth

You can modify the Resource Size to adjust the amount of CPU or GPU cores, RAM and Network Bandwidth. Then you can modify the Number of instances on which your AI Apps will be deployed.

To benefit from High Availability, a minimum of 2 instances are required.

For our first deployment, we will select two instances.

Step 4: Configure your AI App

First, choose a name for your AI Apps or keep the automatically generated name.

Then, you have the option to add some Key/Value labels to filter or organize your AI App access.

As an example, add a label with Key=owner and Value=bastien.

The last parameter to select is the Access rule. Either you restrict your AI Apps access to selected tokens, or you allow any access. Use this option carefully; usually a Public access is used for a test, but not in production since everyone will be able to use it.

We will select Restricted Access for this deployment.

Step 5: Review and launch your AI App

This final step is a summary about your AI App deployment. You can review the previously selected options and parameters. Launch your AI App by clicking on Deploy an AI App

Congratulations, your first AI App is now running on production!

Connect to your first AI App

AI Apps can be your own Docker container or applications proposed by the OVHcloud ecosystem.

In this tutorial, we deployed a HuggingFace AI model allowing us to analyze positive or negative sentiment in English sentences. There is no web interface to manage this AI model. What is given is an API endpoint that you can reach via HTTP.

Step 1: Check your AI Apps status

First, go check your AI App details and verify that your AI App has a running status. Please also verify that your labels are clearly mentioned.

In this example, we added the label owner=bastien and two labels were automatically added by OVHcloud.

Step 2: Generate a security token

During the AI Apps deployment process, we selected "restricted access". To query your AI App, you first need a valid security token.

In your OVHcloud Control Panel, go to your AI Apps list. Select the Token tab.

Click on the New token button then fill in a name, a label selector, a role and region as below:

A few explanations:

• Label selector: you can restrict the token granted by labels. You can note a specific id, a type, or any previously created label such as owner=bastien in our case.
• Role: AI Platform Operator can read and manage your AI App. AI Platform Read only can only read your AI App.
• Region: tokens are regionalized. Select the region related to your AI App.

Generate your first curl query

Now that your AI App is running and token generated, you are ready for your first query.

You can find the cURL example in the bottom part of your AI App, with standard parameters and JSON output format.

Since we are on restricted access, you need to specify the authentication token in the header following this format:

-H "Authorization: Bearer $YOURTOKENHERE"

In our case, the exact cURL code is:

 curl --request POST
    --url https://982a750f-e1f0-45cf-bf41-efc1031f1101.app.bhs.training.ai.cloud.ovh.net/predict
    -H "Authorization: Bearer 0sTWeKaATGuV0a0cqFZeNV2tILbdxgYLoMQRxvWABHYWWDQ4JDs945o4Afhy+6Sj"
    --header 'Content-Type: application/json'
    --data '{
        "inputs": "I love curl so much!"
    }'
 ```

 Giving us 

 ```bash
 {
    "probabilities": [0.003630106, 0.99636996],
    "info": {
        "was_input_truncated": false,
        "num_tokens": 8
    }
}

Probability result is explained in negative sentiment then positive sentiment. Here, we obtain a 99,6% positive score.

Generate your first Python query

If you want to query this model with Python, this code sample with Python Request library may suit you:

import requests
import json
from requests.structures import CaseInsensitiveDict

url = "https://982a750f-e1f0-45cf-bf41-efc1031f1101.app.bhs.training.ai.cloud.ovh.net/predict"

headers = CaseInsensitiveDict()
headers["Authorization"] = "Bearer 0sTWeKaATGuV0a0cqFZeNV2tILbdxgYLoMQRxvWABHYWWDQ4JDs945o4Afhy+6Sj"

data = {'inputs': 'I love curl so much!'}

resp = requests.post(url, headers=headers, json=data)

print(resp.status_code)

print(resp.json())

Result:

200
{'probabilities': [0.003630106, 0.99636996], 'info': {'was_input_truncated': False, 'num_tokens': 8}}

That's it! To go further, you can imagine a script which gathers latest Facebook or Twitter messages and analyze their sentiment over time.

Stop and delete your AI App

You can keep your AI App running as long as you want. At any time, you may stop your application in your AI App details:

Once stopped, your AI App will free up the previously allocated compute resources. Your endpoint is kept and if you restart your AI App, the same endpoint can be reused seamlessly. Also, when you stop your AI App, you don't book compute resources anymore which means you don't have expenses for this part. Only expenses for attached storage may occur.

If you want to completely delete your AI App, just click on the delete action. Be sure to also delete your Object Storage data if you don't need it anymore.

Feedback

Please feel free to send us your questions, feedback and suggestions to help our team improve the service on the OVHcloud Discord server!

These guides might also interest you...