preprocessors=[
"chatterbot.preprocessors.clean_whitespace",
"chatterbot.preprocessors.unescape_html",
"chatterbot.preprocessors.convert_to_ascii"
],
logic_adapters=[{
'import_path': 'chatterbot.logic.BestMatch',
'default_response': 'I am sorry, but I do not understand.',
'maximum_similarity_threshold': 0.90
}, {
'import_path':
'chatterbot.logic.MathematicalEvaluation',
}])
trainer = ListTrainer(bot) # set the trainer
trainer.export_for_training('./my_export.json')
def speak(word):
engine.say(word)
engine.runAndWait()
def Find(query):
# findall() has been used
# with valid conditions for urls in string
url = re.findall(
'http[s]?://(?:[a-zA-Z]|[0-9]|[$-_@.&+]|[!*\(\), ]|(?:%[0-9a-fA-F][0-9a-fA-F]))+',
query)
return url
"We don't have a phone number for this type of account"),
("mailing address",
"You can email our corporate headquaters at [email protected] "),
("chatterbot",
"library making it easy to generate automated responses to a user’s input, visit https://chatterbot.readthedocs.io/en/stable/"
),
("textblob",
"library for processing textual data, please visit https://textblob.readthedocs.io/en/dev/"
)
]
classifier = [
"silly", "dumb", "stupid", "I'dont think so", "I don't care",
"do you know anything", "not good", "omg", "this is not working"
"this is bad", "not what I want", "live help", "get me a rep",
"I need a real person", "forget you"
]
for topic, link in documentation_topics:
trainer.train([f"{topic}", f"Sure, here is the {topic} link: {link}"])
for topic, description in twilio_knowledge:
trainer.train([f"{topic}", f"Ok sure, {description}"])
for i in classifier:
trainer.train([
f"{i}", "I am sorry you feel that way, please ask the question again"
])
trainer.export_for_training('twilybot.json')
database_uri='sqlite:///database.sqlite3')
# For training
trainer = ListTrainer(chatbot)
"""
# Create a new trainer for the chatbot
trainer = ChatterBotCorpusTrainer(chatbot)
#Train the chatbot based on the english corpus only is necessary one time.
trainer.train("chatterbot.corpus.english")
"""
# Select the language (download previously)(sm = small)
nlp = spacy.load("en_core_web_sm")
"""
# We can export the data to a file json
trainer = ChatterBotCorpusTrainer(chatbot)
trainer.export_for_training("./bot_mongo.json")
"""
def speak ():
response = chatbot.get_response(user)
print("BOT: "+str(response))
return response
# With this function we can search in the memory of the bot for check for example the fever of the user.
"""
def reg (txto):
patron = "\d{2}[\.\d]*"
class Bot:
def __init__(self, name, acurrency):
self.name = name
self.acurrency = acurrency
self.bot = ChatBot(name)
# Salva as perguntas e respostas do usuário
self.fulltext = 'Você gosta de cozinhar?'
# Faz o treino em array
def training_array(self, speaking):
self.trainer = ListTrainer(self.bot)
self.trainer.train(speaking)
# Faz o treino em csv
def training_csv(self, path):
csv = CsvFactory(path)
list = csv.to_array()
self.trainer = ListTrainer(self.bot)
self.trainer.train(list)
# Exporta o treino do banco de dados
def export_train(self):
self.trainer.export_for_training('./my_export.json')
def add_model(self, path, rows):
csv = CsvFactory(path)
csv.write(rows)
#### Acuracia Naive Bayes #####
def get_accuracy(self, path):
dataset = pd.read_csv(path, sep=';', encoding="ISO-8859-1")
token = RegexpTokenizer(r'[a-zA-Z0-9]+')
cv = CountVectorizer(analyzer='word',
lowercase=True,
stop_words='english',
min_df=1,
ngram_range=(1, 1),
tokenizer=token.tokenize)
text_counts = cv.fit_transform(dataset['Text'])
# Aqui começa o treinamento
y = dataset['Sentiment']
X = text_counts
X_train, X_test, y_train, y_test = train_test_split(X,
y,
test_size=0.3,
random_state=10)
NaiveB = MultinomialNB()
NaiveB.fit(X_train, y_train)
y_pred_test_NaiveB = NaiveB.predict(X_test)
#retorno
print('Acurácia NaiveBayes:',
metrics.accuracy_score(y_test, y_pred_test_NaiveB))
# Resposta do robo
def response(self, question):
answer = self.bot.get_response(question)
text = 'Eita, não entendi.'
if float(answer.confidence) > self.acurrency:
# acumulador para salvar no futuro modelo
self.fulltext += ' ' + question + ' ' + str(answer)
text = str(answer)
return text
if (joke['rating'] > 3 and len(joke['body']) < 70
and joke['category'] != 'Blonde Jokes'):
chosenJokes.append("Tell me a joke")
chosenJokes.append(joke['body'])
count += 1
# with open('wocka.json', encoding="utf-8") as json_file:
# data = json.load(json_file)
# for joke in data:
# if (len(joke['body']) < 70 and not ('Blond' in joke['category']) ):
# chosenJokes.append( re.sub(r'[^\x00-\x7F]+',' ', joke['body'].replace('\r', "").replace('\n', "")))
# count+=1
trainer.train(chosenJokes)
trainer.export_for_training('./full_list_trainer.json')
# # The following loop will execute each time the user enters input
# while True:
# try:
# user_input = input(">> ")
# bot_response = chatbot.get_response(user_input)
# print(bot_response)
# # Press ctrl-c or ctrl-d on the keyboard to exit
# except (KeyboardInterrupt, EOFError, SystemExit):
# break
}
],
database_uri='sqlite:///database-chatbot.db' #use sqlite
#database_uri = "postgresql://*****:*****@localhost/rus_bot" #use postgres
)
# Create a new trainer for the chatbot, using corpus trainer - uncomment, if demand
trainer = ListTrainer(bot)
# Train the chatbot based on the english corpus - uncomment, if demand
#trainer.train("chatterbot.corpus.english")
#print('Type something to begin...')
for dialog in train_corpus_full:
trainer.train(dialog)
trainer.export_for_training('./attempt1.json')
# if __name__ =='__main__':
# # The following loop will execute each time the user enters input
# trainer.train(train_corpus_full)
# while True:
# try:
# user_input = input()
# bot_response = bot.get_response(user_input)
# print(bot_response)
# # Press ctrl-c or ctrl-d on the keyboard to exit
# except (KeyboardInterrupt, EOFError, SystemExit):
"isis", "taliban", "cp", "bomb", "ied", "explosive"
] #more bad words to limit searches, cos I really don't trust people
f = open(f"{filepath}/config/token.txt")
token = str(
f.readline()
) #Gets the token from another text file so I don't have to leave the token in this file where anyone can read it
f.close()
null = None #so I can write "null" instead of "None" and look like hackerman
#Chatterbot stuff
chatbot = ChatBot('Bottombot')
trainer = ListTrainer(chatbot)
trainer.export_for_training(
f'{filepath}/my_export.json') #exporting the saved training data
trainer.train([
"UwU" #It starts with one word. I chose this
])
#Chatterbot stuff
client = discord.Client()
client = commands.Bot(command_prefix="-")
client.remove_command('help')
starttime = null
bannedids = [
369234715015905292, 567522840752947210, 459685744387162112,
