CODEWITHSUNDEEP
google-cloud-vertex-ai
Sanyo Mn
Sanyo Mn

Reputation: 431

Sending http request Google Vertex AI end point

I've just deployed an ML model on Google vertex AI, it can make predictions using vertex AI web interface. But is it possible to send a request from a browser, for example, to this deployed model. Something like

http://myapp.cloud.google.com/input="features of an example"

and get the prediction as output. Thanks

Upvotes: 5

Views: 10766

Answers (3)

Saripudin
Saripudin

Reputation: 72

Another alternatif you can use Bearer Token to get prediction

  1. Install gcloud CLI. Reference
  2. Run gcloud auth application-default login and you must login your account
  3. Run gcloud auth print-access-token to get the token

Now you can hit the endpoint by using cURL command or via Postman

curl \
-X POST \
-H "Authorization: Bearer $(gcloud auth print-access-token)" \
-H "Content-Type: application/json" \
https://us-central1-aiplatform.googleapis.com/ui/projects/${PROJECT_ID}/locations/us-central1/endpoints/${ENDPOINT_ID}:predict -d '{
  "instances": {
    "mimeType": "text/plain",
    "content": "YOUR_TEXT_CONTENT"
  }
}'

Upvotes: 1

mon
mon

Reputation: 22234

Endpoint URL

Vertex AI document - guide - Send an online prediction request

HTTP method and URL:

  • LOCATION: The region where you are using Vertex AI.
  • PROJECT: Your project ID
  • ENDPOINT_ID: The ID for the endpoint.
POST https://LOCATION-aiplatform.googleapis.com/v1/projects/PROJECT/locations/LOCATION/endpoints/ENDPOINT_ID:predict

Alternatively we can use gcloud ai endpoints list command.

$ LOCATION='us-central1'
$ gcloud ai endpoints list --region=${LOCATION} --uri
Using endpoint [https://us-central1-aiplatform.googleapis.com/]
https://us-central1-aiplatform.googleapis.com/v1beta1/projects/335513839215/locations/us-central1/endpoints/5031620295400620032

We can check the available Vertex AI API service endpoint URL FQDN part in Vertex AI API - Service: aiplatform.googleapis.com

https://us-central1-aiplatform.googleapis.com
https://us-east1-aiplatform.googleapis.com
https://us-east4-aiplatform.googleapis.com
https://us-west1-aiplatform.googleapis.com
https://us-west2-aiplatform.googleapis.com
https://northamerica-northeast1-aiplatform.googleapis.com
https://northamerica-northeast2-aiplatform.googleapis.com
https://europe-west1-aiplatform.googleapis.com
https://europe-west2-aiplatform.googleapis.com
https://europe-west3-aiplatform.googleapis.com
https://europe-west4-aiplatform.googleapis.com
https://europe-west6-aiplatform.googleapis.com
https://asia-east1-aiplatform.googleapis.com
https://asia-east2-aiplatform.googleapis.com
https://asia-northeast1-aiplatform.googleapis.com
https://asia-northeast3-aiplatform.googleapis.com
https://asia-south1-aiplatform.googleapis.com
https://asia-southeast1-aiplatform.googleapis.com
https://australia-southeast1-aiplatform.googleapis.com

Google API for Vertex endpoint prediction

The URL format above is based on the API definition Method: projects.locations.endpoints.predict.

POST https://{service-endpoint}/v1/{endpoint}:predict

Where {service-endpoint} is one of the supported service endpoints.

Authentication

GCP uses OAuth 2.0 for authentication by default, and the Vertex AI endpoint authenticates the access with the OAuth 2.0 access token set in the HTTP Authorization header as a Bearer token.

RFC 6749 - The OAuth 2.0 Authorization Framework

     +--------+                               +---------------+
     |        |--(A)- Authorization Request ->|   Resource    |
     |        |                               |     Owner     |
     |        |<-(B)-- Authorization Grant ---|               |
     |        |                               +---------------+
     |        |
     |        |                               +---------------+
     |        |--(C)-- Authorization Grant -->| Authorization |
     | Client |                               |     Server    |
     |        |<-(D)----- Access Token -------|               |
     |        |                               +---------------+
     |        |
     |        |                               +---------------+
     |        |--(E)----- Access Token ------>|    Resource   |
     |        |                               |     Server    |
     |        |<-(F)--- Protected Resource ---|               |
     +--------+                               +---------------+

If we have gcloud command available e.g. inside a GCP environment, we can use the ADC (Application Default Credentials) to first get authenticated with gcloud auth application-default login and obtain the token using the gloud auth print-access-token command. Then the token can be specified to the HTTP header as in the Vertex AI documentation below.

curl -X POST \
-H "Authorization: Bearer "$(gcloud auth application-default print-access-token) \
-H "Content-Type: application/json; charset=utf-8" \
-d @request.json \
"https://LOCATION-aiplatform.googleapis.com/v1/projects/PROJECT/locations/LOCATION/endpoints/ENDPOINT_ID:predict"

Request body format

{
  "instances": [
    value
  ],
  "parameters": value
}

The actual value to set depends on the frameworks. See Get online predictions from custom-trained models- Format your input for online prediction if you are getting predictions from custom models.

Upvotes: 1

Amal Sarkar
Amal Sarkar

Reputation: 122

Yes, you can send using endpoint URL as.

https://us-central1-aiplatform.googleapis.com/v1beta1/projects/<PROJECT_ID>/locations/us-central1/endpoints/<ENDPOINT_ID>:predict

Data should be given as in POST parameter.

{
  "instances": 
    [1.4838871833555929,
 1.8659883497083019,
 2.234620276849616,
 1.0187816540094903,
 -2.530890710602246,
 -1.6046416850441676,
 -0.4651483719733302,
 -0.4952254087173721,
 0.774676376873553]
}

URL should be Region Based.

Upvotes: 7

Related Questions