def run_bot_online(interpreter,
domain_file="domain.yml",
training_data_file='data/stories.md'):
'''
This function trains the bot in an interactive manner
:param interpreter: NLU Interpreter
:param domain_file: Domain file
:param training_data_file: Chat story board file
:return: Agent
'''
action_endpoint = EndpointConfig(url="http://localhost:5055/webhook")
agent = Agent(domain_file,
policies=[
MemoizationPolicy(max_history=3),
KerasPolicy(max_history=5, epochs=300, batch_size=50)
],
interpreter=interpreter,
action_endpoint=action_endpoint)
data = agent.load_data(training_data_file)
agent.train(data)
interactive.run_interactive_learning(agent, training_data_file)
return agent
def run_weather_online(interpreter, domain_file="weather_domain.yml", training_data_file = './data/stories.md'): action_endpoint = EndpointConfig(url = "http://localhost:5005/webhook") agent = Agent(domain_file, policies = [MemoizationPolicy(max_history = 2), KerasPolicy(max_history = 3, epochs = 3, batch_size = 50)], interpreter = interpreter, action_endpoint = action_endpoint) training_data = agent.load_data(training_data_file) agent.train(training_data) interactive.run_interactive_learning(agent, training_data_file) return agent
def train_interactive():
current_directory = os.path.dirname(os.path.realpath(__file__))
domain_file = 'domain.yml'
interpreter = RasaNLUInterpreter(current_directory + '/agent-data/models/nlu/default/current')
action_endpoint = EndpointConfig(url=actionIP)
stories_file = current_directory + '/agent-data/data/stories.md'
agent = Agent(domain_file,
policies=[MemoizationPolicy(max_history=5), KerasPolicy()],
interpreter=interpreter, action_endpoint=action_endpoint)
data = agent.load_data(stories_file)
agent.train(data)
interactive.run_interactive_learning(agent, stories_file)
return agent
def run_laliga_online(interpreter,
domain_file="./LaLiga_bot/Laliga_domain.yml",
training_data_file='./LaLiga_bot/data/stories.md'):
action_endpoint = EndpointConfig(url="http://localhost:5055/webhook")
'''
train from scratch
agent = Agent(domain_file,
policies=[MemoizationPolicy(max_history=2), KerasPolicy(max_history=3, epochs=3, batch_size=50)],
interpreter=interpreter,
action_endpoint=action_endpoint)
data = agent.load_data(training_data_file)
agent.train(data)
'''
agent = Agent.load(r'.\LaLiga_bot\models\dialogue', interpreter=interpreter, action_endpoint=action_endpoint)
interactive.run_interactive_learning(agent, training_data_file)
return agent
def run_chatbot_online(interpreter,
domain_file="chatbot_domain.yml",
training_data_file='data/stories.md'):
action_endpoint = EndpointConfig(url="http://localhost:5055/webhook")
agent = Agent(domain_file,
policies=[
MemoizationPolicy(max_history=2),
KerasPolicy(max_history=3, epochs=3, batch_size=90)
],
interpreter=interpreter,
action_endpoint=action_endpoint)
#max history stands for number of states our bot should remember, epochs is number of forward/backward passes of traning in process,batch size is amount of traning examples should be used
data = agent.load_data(training_data_file)
agent.train(data)
interactive.run_interactive_learning(agent,
training_data_file,
skip_visualization=True)
return agent
def run_weather_online(interpreter,
domain_file="domain.yml",
training_data_file='data/stories.md'):
action_endpoint = EndpointConfig(url="http://localhost:5055/webhook")
agent = Agent(domain_file,
policies=[
MemoizationPolicy(max_history=5),
KerasPolicy(max_history=5, epochs=4, batch_size=70),
FormPolicy()
],
interpreter=interpreter,
action_endpoint=action_endpoint)
data = agent.load_data(training_data_file)
agent.train(data)
interactive.run_interactive_learning(agent,
training_data_file,
skip_visualization=True)
return agent
def run_online(interpreter,
domain_file="domain.yml",
training_data_file='data/stories.md'):
action_endpoint = EndpointConfig(url="http://localhost:5055/webhook")
model_path = './models/dialogue' # Where to save the model after its trained
agent = Agent(domain_file,
policies=[
MemoizationPolicy(),
KerasPolicy(max_history=3, epochs=100, batch_size=50)
],
interpreter=interpreter,
action_endpoint=action_endpoint)
data = agent.load_data(training_data_file, augmentation_factor=50)
agent.train(data)
agent.persist(model_path) # Save the Model
interactive.run_interactive_learning(agent,
training_data_file,
skip_visualization=True)
return agent
def run_cbeg(interpreter,
domain_file="cbeg_domain.yml",
training_data_file='data/stories.md'):
action_endpoint = EndpointConfig(url="http://localhost:5055/webhook")
fallback = FallbackPolicy(fallback_action_name="action_default_fallback",
core_threshold=0.4,
nlu_threshold=0.4)
agent = Agent(domain_file,
policies=[
MemoizationPolicy(max_history=2),
KerasPolicy(max_history=3, epochs=3, batch_size=50)
],
interpreter=interpreter,
action_endpoint=action_endpoint)
data = agent.load_data(training_data_file)
agent.train(data)
interactive.run_interactive_learning(agent, training_data_file)
return agent
def run_bot_online(interpreter, domain_file, training_data_file):
action_endpoint = EndpointConfig(url="http://localhost:5055/webhook")
fallback = FallbackPolicy(fallback_action_name="action_default_fallback",
core_threshold=0.3,
nlu_threshold=0.3)
agent = Agent(domain=domain_file,
policies=[
MemoizationPolicy(max_history=6),
KerasPolicy(max_history=6, epochs=200), fallback,
FormPolicy()
],
interpreter=interpreter,
action_endpoint=action_endpoint)
data = agent.load_data(training_data_file)
agent.train(data)
interactive.run_interactive_learning(agent, training_data_file)
return agent
def run_trade_online(interpreter,
domain_file=Path.TRADE_DOMAIN.value,
training_data_file=Path.STORIES.value):
fallback = FallbackPolicy(fallback_action_name="action_default_fallback",
core_threshold=0.5,
nlu_threshold=0.5)
action_endpoint = EndpointConfig(url="http://localhost:5055/webhook")
agent = Agent(domain_file,
policies=[
MemoizationPolicy(max_history=2),
KerasPolicy(max_history=3, epochs=3, batch_size=50),
fallback
],
interpreter=interpreter,
action_endpoint=action_endpoint)
data = agent.load_data(training_data_file)
agent.train(data)
interactive.run_interactive_learning(agent, training_data_file)
return agent
def run_recipe_online(interpreter,
domain_file="recipes_domain.yml",
training_data_file="data/stories.md"):
# Set up endpoint for Custom Action
# NOTE: Custom Actions run on a separate server
action_endpoint = EndpointConfig(url="http://localhost:5055/webhook")
# Create Agent
# NOTE: Agent processes stories and uses Rasa Interpreter for input analysis
agent = Agent(domain_file,
policies=[
MemoizationPolicy(max_history=5),
KerasPolicy(max_history=5, epochs=6, batch_size=150)
],
interpreter=interpreter,
action_endpoint=action_endpoint)
# Train agent
data = agent.load_data(training_data_file)
agent.train(data)
interactive.run_interactive_learning(agent, training_data_file)
return agent
