Training Dialogflow Agent using Python SDK

Question

i using dialogflow v2 with python sdk.

All works fine, except when i add a new intent with its training phrases. The bot doesnt recognize the phrases until i enter through the web console and save the intent, when the training start, after that, the bot works well.

I tried to train the intent using the python sdk:

agent_client = dialogflow.AgentsClient(credentials=self.credentials)
response = agent_client.train_agent('projects/' + self.project_id)

The response is 200, but the agent wasnt trained.

Thanks to any clue how to make this work.

I'm asking because the SDKs methods are asynchronous, train_agent returns a future which does not initially contain any information about the result of the training operation. — gmolau
– gmolau, Commented Mar 15, 2019 at 20:23

v2a · Accepted Answer · 2021-05-25 20:50:53Z

1

May be, it helps:

def train_agent(project_id):
    from google.cloud import dialogflow

    agents_client = dialogflow.AgentsClient()
    parent = dialogflow.AgentsClient.common_project_path(project_id)
    response = agents_client.train_agent(
        request={"parent": parent}
    )

    print(response.done())

P.S.: GOOGLE_APPLICATION_CREDENTIALS in my .env

answered May 25, 2021 at 20:50

v2a

111 bronze badge

Sign up to request clarification or add additional context in comments.

Comments

Ivan Aguas · Accepted Answer · 2022-08-17 16:15:24Z

0

This works for me. Currently, I'm using FastApi. The secret is to add the function done(). That would return true or false if the operation finished successfully.

@router.post("/chatbot/train/{project_id}", response_model=IGetResponseBase)
async def train_agent(
    project_id: str, 
) -> Any:

# Create a client
agent_client = AgentsAsyncClient()
parent = agent_client.common_project_path(project_id)

# Initialize request argument(s)
request = TrainAgentRequest(
    parent=parent,
)

# Make the request
operation = await agent_client.train_agent(request=request)

print("Waiting for operation to complete...")


response = await operation.done()

# Handle the response
print('operation', response)

answered Aug 17, 2022 at 16:15

Ivan Aguas

572 bronze badges

Collectives™ on Stack Overflow

Training Dialogflow Agent using Python SDK

2 Answers 2

Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related