How to download bulk training phrases from DialogFlow into spreadsheet - dialogflow-es

Question

I am a beginner to Dialog flow .I want to know if there is a way to download bulk training phrases from DialogFlow into spreadsheet

Answer 1

Dialogflow does not have a feature that could bulk download the training phrases only. You can submit a feature request for this since it seems that it will be really useful.
You can only get the training phrases if you parse get_intent(). Using the code below (Python) I generated a CSV file that has the training phrases per intent name you have defined in the code.
def list_training_phrases(project_id, intent_arr):
"""Returns all training phrases for a specified intent."""
from google.cloud import dialogflow
from google.protobuf.json_format import MessageToDict
import csv
# Create the intents client
intent_client = dialogflow.IntentsClient()
# The options for views of an intent
intent_view = dialogflow.IntentView.INTENT_VIEW_FULL
# Compose the get-intent request
csv_arr = []
for intent_name in intent_arr:
get_intent_request = dialogflow.GetIntentRequest(
name=intent_name, intent_view=intent_view
)
intent = intent_client.get_intent(get_intent_request)
intent_dict = MessageToDict(intent._pb) # convert response to dictionary
phrases_arr = []
fixed_arr = []
# Iterate through the training phrases.
for phrases in intent_dict["trainingPhrases"]:
phrases_arr.append(phrases["parts"])
for entry in phrases_arr:
new_dict = merge_list_of_dictionaries(entry)
fixed_arr.append( ''.join(new_dict["text"]))
print(fixed_arr)
csv_arr.append(fixed_arr)
with open('./train_phrases.csv', 'w', encoding='UTF8') as f:
writer = csv.writer(f)
for row in csv_arr:
writer.writerow(row)
# [END dialogflow_list_training_phrases]
# https://stackoverflow.com/questions/45649141/combine-values-of-same-keys-in-a-list-of-dicts
def merge_list_of_dictionaries(dict_list):
new_dict = {}
for d in dict_list:
for d_key in d:
if d_key not in new_dict:
new_dict[d_key] = []
new_dict[d_key].append(d[d_key])
return new_dict
def list_intents(project_id):
from google.cloud import dialogflow
intents_client = dialogflow.IntentsClient()
parent = dialogflow.AgentsClient.agent_path(project_id)
intents = intents_client.list_intents(request={"parent": parent})
intent_arr = []
intent_to_check = ["Make Appointment", "Hours"] # define your intents here
for intent in intents:
if intent.display_name in intent_to_check:
intent_arr.append(intent.name)
list_training_phrases(project_id=project_id, intent_arr=intent_arr)
project_id = "your-project-id"
list_intents(project_id = project_id)
Per line of the output file represents the training phrases per intent. Line 1, represents "Hours" intent and Line 2 as "Make Appointment" intent.
See output below (train_phrases.csv):