from app.nlu.entity_extractor import EntityExtractor
from app.intents.models import Intent
from app import app
from app.nlu.intent_classifer import IntentClassifier
from app import my_signals
model_updated_signal = my_signals.signal('model-updated')
def train_models():
"""
Initiate NER and Intent Classification training
:return:
"""
# generate intent classifier training data
intents = Intent.objects
if not intents:
raise Exception("NO_DATA")
# train intent classifier on all intents
train_intent_classifier(intents)
# train ner model for each Stories
for intent in intents:
train_all_ner(str(intent.id), intent.trainingData)
model_updated_signal.send(app, message="Training Completed.")
from app.nlu.entity_extractor import EntityExtractor
from app.intents.models import Intent
from app import app
from app.nlu.classifiers.starspace_intent_classifier import EmbeddingIntentClassifier
from app import my_signals
model_updated_signal = my_signals.signal('model-updated')
def train_models():
"""
Initiate NER and Intent Classification training
:return:
"""
# generate intent classifier training data
intents = Intent.objects
if not intents:
raise Exception("NO_DATA")
# train intent classifier on all intents
train_intent_classifier(intents)
# train ner model for each Stories
for intent in intents:
train_all_ner(str(intent.intentId), intent.trainingData)
model_updated_signal.send(app,message="Training Completed.")
def train_intent_classifier(intents):