I trained a custom model using Google Cloud AutoML. Now i am trying to access it, using the script provided by Google.
I tried to vary "content" in any kind of ways. I also had a look at the information provided here. Surely i did provide the correct path to the key file. Also i checked on the project ID and model ID. I do have a service account. Billing is enabled too.
export GOOGLE_APPLICATION_CREDENTIALS={key-file-path}
curl -X POST \
-H "Authorization: Bearer $(gcloud auth application-default print-access-token)" \
-H "Content-Type: application/json" \
https://automl.googleapis.com/v1beta1/projects/{my-project-ID}/locations/us-central1/models/{my-model-id}:predict \
-d '{
"payload" : {
"textSnippet": {
"content": "happy",
"mime_type": "text/plain"
},
}
}'
I expect the result to be the prediction.
My result looks like this:
"error": {
"code": 400,
"message": "Invalid JSON payload received. Expected a value.\n“happy”,\n \n^",
"status": "INVALID_ARGUMENT"}