IssueTracker

Using Cloud Shell Terminal, there are multiple errors being returned after you compile the auto-generated Python script, shown at the 'Test & Use' tab when you want to try out the code for your trained model in Custom Use.

Problems I have encountered:
1. There is a Module-error when compiling:
line 5, in <module>
    from google.cloud.automl_v1 import service_pb
ImportError: cannot import name 'service_pb' from 'google.cloud.automl_v1' (/home/laurensvreekamp/.local/lib/python3.7/site-packages/google/cloud/automl_v1/__init__.py)

2. There is an indent too much on the very last line (" print"), that results in an error using Python 3.

3. The first argument used when calling the prediction-function (on last line) is called 'content' but should be named 'file_path'

4. After fixing all this, having installed all proper all modules/packages, setting service account credentials and everything, I get this error-code, using Cloud Shell, terminal and editor:

Finally after fixing these errors, and uncommenting the import of the module on line 5
Traceback (most recent call last):
  File "predict.py", line 30, in <module>
    print(get_prediction(file_path, model_name))
  File "predict.py", line 24, in get_prediction
    request = prediction_client.predict(model_name, payload, params)
TypeError: predict() takes from 1 to 2 positional arguments but 4 were given

Steps to reproduce:
Open Cloud Shell Terminal
Install necessary packages/modules
Create a new file: 1. Predict.py
Create a txt file: as the source to analyse by the model for classification
Get the id for the trained model

Run script in the file 'predict.py'

Here's the current script:
---
import sys

from google.api_core.client_options import ClientOptions
from google.cloud import automl_v1
#from google.cloud.automl_v1 import service_pb

def inline_text_payload(file_path):
  with open(file_path, 'rb') as ff:
    content = ff.read()
  return {'text_snippet': {'content': content, 'mime_type': 'text/plain'} }

def pdf_payload(file_path):
  return {'document': {'input_config': {'gcs_source': {'input_uris': [file_path] } } } }

def get_prediction(file_path, model_name):
  options = ClientOptions(api_endpoint='automl.googleapis.com')
  prediction_client = automl_v1.PredictionServiceClient(client_options=options)

  payload = inline_text_payload(file_path)
  # Uncomment the following line (and comment the above line) if want to predict on PDFs.
  #payload = pdf_payload(file_path)

  params = {}
  request = prediction_client.predict(model_name, payload, params)
  return request  # waits until request is returned

if __name__ == '__main__':
  file_path = sys.argv[1]
  model_name = sys.argv[2]
  print(get_prediction(file_path, model_name))
---

What you expected to happen:
Returning two labels -the classification of my text- and a prediction score for both labels

Other information (workarounds you have tried, documentation consulted, etc):
1. I've read through all the documentation on Cloud and the AutoML API, but no clear help found there.
2. Looked to Stack-overflow and found someone who had similar problems, but not my fourth issue/errora
3. Browsed GitHub to find similar issues and questions, alternative solutions, but non were satisfactory addressing this issue