3

Look like I am the first poster here. I am an app developer on Android/iOS and have been waiting to see Cloud AutoML Vision for public release. I just play around the console and build a model with 2 labels. However, the implementation require Rest API or Python , both of which I am not familiar with. How do I make use of of built model? Any other GCP product do I need to use?

how or where do I execute the REST API?

request.json

{
  "payload": {
    "image": {
      "imageBytes": "YOUR_IMAGE_BYTE"
    },
  }
}

Execute the request

curl -X POST -H "Content-Type: application/json" \
  -H "Authorization: Bearer $(gcloud auth application-default print-access-token)" \
  https://automl.googleapis.com/v1beta1/projects/drughk-9747d/locations/us-central1/models/ICN6566958205491349320:predict -d @request.json

I noticed Firebase ML kit where you can upload a .tflite model to Firebase Console. Could Cloud AutoML Vision export the model as a tensorflow or tensorflow lite model? Do Cloud AutoML Vision actually use tensorflow? I really hope Firebase ML kit will implement similar functionality to Cloud AutoML Vision for easier mobile app developement

Ibrahim Ulukaya
  • 12,767
  • 1
  • 33
  • 36
Simon Ho
  • 588
  • 1
  • 8
  • 21

1 Answers1

0

You can call it using Firebase Cloud Functions and alpha version of Google's node.js client lib. Docs could be found here: https://cloud.google.com/nodejs/docs/reference/automl/0.1.x/v1beta1.PredictionServiceClient

lexotrion
  • 84
  • 1
  • 4