def validate(self):
# type: () -> None
"""Ensures that the loaded training data is valid, e.g. has a minimum of certain training examples."""
logger.debug("Validating training data...")
examples = self.sorted_intent_examples()
different_intents = []
for intent, group in groupby(examples, lambda e: e.get("intent")):
size = len(list(group))
different_intents.append(intent)
if size < self.MIN_EXAMPLES_PER_INTENT:
template = "Intent '{}' has only {} training examples! minimum is {}, training may fail."
warnings.warn(template.format(intent, size, self.MIN_EXAMPLES_PER_INTENT))
different_entities = []
for entity, group in groupby(self.sorted_entity_examples(), lambda e: e["entity"]):
size = len(list(group))
different_entities.append(entity)
if size < self.MIN_EXAMPLES_PER_ENTITY:
template = "Entity '{}' has only {} training examples! minimum is {}, training may fail."
warnings.warn(template.format(entity, size, self.MIN_EXAMPLES_PER_ENTITY))
logger.info("Training data stats: \n" +
"\t- intent examples: {} ({} distinct intents)\n".format(
self.num_intent_examples, len(different_intents)) +
"\t- found intents: {}\n".format(list_to_str(different_intents)) +
"\t- entity examples: {} ({} distinct entities)\n".format(
self.num_entity_examples, len(different_entities)) +
"\t- found entities: {}\n".format(list_to_str(different_entities)))
def validate(self):
# type: () -> None
"""Ensures that the loaded training data is valid, e.g. has a minimum of certain training examples."""
logger.debug("Validating training data...")
examples = self.sorted_intent_examples()
different_intents = []
for intent, group in groupby(examples, lambda e: e.get("intent")):
size = len(list(group))
different_intents.append(intent)
if size < self.MIN_EXAMPLES_PER_INTENT:
template = "Intent '{}' has only {} training examples! minimum is {}, training may fail."
warnings.warn(template.format(intent, size, self.MIN_EXAMPLES_PER_INTENT))
different_entities = []
for entity, group in groupby(self.sorted_entity_examples(), lambda e: e["entity"]):
size = len(list(group))
different_entities.append(entity)
if size < self.MIN_EXAMPLES_PER_ENTITY:
template = "Entity '{}' has only {} training examples! minimum is {}, training may fail."
warnings.warn(template.format(entity, size, self.MIN_EXAMPLES_PER_ENTITY))
logger.info("Training data stats: \n" +
"\t- intent examples: {} ({} distinct intents)\n".format(
self.num_intent_examples, len(different_intents)) +
"\t- found intents: {}\n".format(list_to_str(different_intents)) +
"\t- entity examples: {} ({} distinct entities)\n".format(
self.num_entity_examples, len(different_entities)) +
"\t- found entities: {}\n".format(list_to_str(different_entities)))
def print_stats(self):
logger.info(
"Training data stats: \n" +
"\t- intent examples: {} ({} distinct intents)\n".format(
len(self.intent_examples), len(self.intents)) +
"\t- Found intents: {}\n".format(list_to_str(self.intents)) +
"\t- entity examples: {} ({} distinct entities)\n".format(
len(self.entity_examples), len(self.entities)) +
"\t- found entities: {}\n".format(list_to_str(self.entities)))
def print_stats(self):
logger.info("Training data stats: \n" +
"\t- intent examples: {} ({} distinct intents)\n".format(
len(self.intent_examples), len(self.intents)) +
"\t- Found intents: {}\n".format(
list_to_str(self.intents)) +
"\t- entity examples: {} ({} distinct entities)\n".format(
len(self.entity_examples), len(self.entities)) +
"\t- found entities: {}\n".format(
list_to_str(self.entities)))
