Guia passo a passo para criar um chatbot personalizado utilizando spaCy (livraria Python NLP)

Olá Comunidade,

Nesse artigo, demonstrarei os seguintes passos para criar seu próprio chatbot utilizando spaCy (spaCy é uma biblioteca de software de código aberto para o processamento avançado de linguagem natural, escrita nas linguagens de programação Python e Cython):

• Passo 1: Instalar as livrarias necessárias

• Passo 2: Criar o arquivo de padrões e respostas

• Passo 3: Treinar o modelo

• Passo 4: Criar uma aplicação ChatBot baseada no modelo treinado

Comecemos

Passo 1: Instalar as livrarias necessárias 

Em primeiro lugar, precisamos instalar as livrarias python necessárias, o que conseguimos executando o seguinte comando:

pip3 install spacy nltk
pip3 install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cpu

Passo 2: Criar o arquivo de padrões e respostas

Necessitamos criar um arquivo intents.json que contenha padrões de perguntas e respostas. A seguir se mostra um exemplo de alguns dos padrões de perguntas e respostas

{
  "intents": [
    {
      "tag": "greeting",
      "patterns": [
        "Hi",
        "Hey",
        "How are you",
        "Is anyone there?",
        "Hello",
        "Good day"
      ],
      "responses": [
        "Hey :-)",
        "Hello, thanks for visiting",
        "Hi there, what can I do for you?",
        "Hi there, how can I help?"
      ]
    },
    {
      "tag": "goodbye",
      "patterns": ["Bye", "See you later", "Goodbye"],
      "responses": [
        "See you later, thanks for visiting",
        "Have a nice day",
        "Bye! Come back again soon."
      ]
    },
    {
      "tag": "thanks",
      "patterns": ["Thanks", "Thank you", "That's helpful", "Thank's a lot!"],
      "responses": ["Happy to help!", "Any time!", "My pleasure"]
    },
    {
      "tag": "items",
      "patterns": [
        "tell me about this app",
        "What kinds of technology used?",
        "What do you have?"
      ],
      "responses": [
        "Write something about the app."
      ]
    },        
    {
      "tag": "funny",
      "patterns": [
        "Tell me a joke!",
        "Tell me something funny!",
        "Do you know a joke?"
      ],
      "responses": [
        "Why did the hipster burn his mouth? He drank the coffee before it was cool.",
        "What did the buffalo say when his son left for college? Bison."
      ]
    }
  ]
}

Passo 3: Treinar o modelo

3.1- Criar um arquivo NeuralNet model.py que se utilizará para treinar o modelo

#model.py file
import torch.nn as nn

class NeuralNet(nn.Module):
    def __init__(self, input_size, hidden_size, num_classes):
        super(NeuralNet, self).__init__()
        self.l1 = nn.Linear(input_size, hidden_size) 
        self.l2 = nn.Linear(hidden_size, hidden_size) 
        self.l3 = nn.Linear(hidden_size, num_classes)
        self.relu = nn.ReLU()
    
    def forward(self, x):
        out = self.l1(x)
        out = self.relu(out)
        out = self.l2(out)
        out = self.relu(out)
        out = self.l3(out)
        # no activation and no softmax at the end
        return out

3.2 - Precisamos do arquivo nltk_utils.py, que utiliza a livraria python «nltk» (uma plataforma utilizada para construir programas Python, que trabalham com dados de linguagem humana, para aplicá-los no processamento estatístico de linguagem natural (PLN)). 

#nltk_utils.py
#nltk library used for building Python programs that work with human language data for applying in 
#statistical natural language processing (NLP)
import numpy as np
import nltk
from nltk.stem.porter import PorterStemmer
stemmer = PorterStemmer()

def tokenize(sentence):
    """
    split sentence into array of words/tokens
    a token can be a word or punctuation character, or number
    """
    return nltk.word_tokenize(sentence)


def stem(word):
    """
    stemming = find the root form of the word
    examples:
    words = ["organize", "organizes", "organizing"]
    words = [stem(w) for w in words]
    -> ["organ", "organ", "organ"]
    """
    return stemmer.stem(word.lower())


def bag_of_words(tokenized_sentence, words):
    """
    return bag of words array:
    1 for each known word that exists in the sentence, 0 otherwise
    example:
    sentence = ["hello", "how", "are", "you"]
    words = ["hi", "hello", "I", "you", "bye", "thank", "cool"]
    bog   = [  0 ,    1 ,    0 ,   1 ,    0 ,    0 ,      0]
    """
    # stem each word
    sentence_words = [stem(word) for word in tokenized_sentence]
    # initialize bag with 0 for each word
    bag = np.zeros(len(words), dtype=np.float32)
    for idx, w in enumerate(words):
        if w in sentence_words: 
            bag[idx] = 1
    return bag

3.3 - Temos que treinar o modelo. Vamos utilizar o arquivo train.py para criá-lo.

#train.py
import numpy as np
import json,torch
import torch.nn as nn
from torch.utils.data import Dataset, DataLoader
from nltk_utils import bag_of_words, tokenize, stem
from model import NeuralNet
 
with open('intents.json', 'r') as f:
    intents = json.load(f)

all_words = []
tags = []
xy = []
# loop through each sentence in our intents patterns
for intent in intents['intents']:
    tag = intent['tag']
    # add to tag list
    tags.append(tag)
    for pattern in intent['patterns']:
        # tokenize each word in the sentence
        w = tokenize(pattern)
        # add to our words list
        all_words.extend(w)
        # add to xy pair
        xy.append((w, tag))

# stem and lower each word
ignore_words = ['?', '.', '!']
all_words = [stem(w) for w in all_words if w not in ignore_words]
# remove duplicates and sort
all_words = sorted(set(all_words))
tags = sorted(set(tags))

print(len(xy), "patterns")
print(len(tags), "tags:", tags)
print(len(all_words), "unique stemmed words:", all_words)

# create training data
X_train = []
y_train = []
for (pattern_sentence, tag) in xy:
    # X: bag of words for each pattern_sentence
    bag = bag_of_words(pattern_sentence, all_words)
    X_train.append(bag)
    # y: PyTorch CrossEntropyLoss needs only class labels, not one-hot
    label = tags.index(tag)
    y_train.append(label)

X_train = np.array(X_train)
y_train = np.array(y_train)

# Hyper-parameters 
num_epochs = 1000
batch_size = 8
learning_rate = 0.001
input_size = len(X_train[0])
hidden_size = 8
output_size = len(tags)
print(input_size, output_size)

class ChatDataset(Dataset):
    def __init__(self):
        self.n_samples = len(X_train)
        self.x_data = X_train
        self.y_data = y_train

    # support indexing such that dataset[i] can be used to get i-th sample
    def __getitem__(self, index):
        return self.x_data[index], self.y_data[index]

    # we can call len(dataset) to return the size
    def __len__(self):
        return self.n_samples

dataset = ChatDataset()
train_loader = DataLoader(dataset=dataset,
                          batch_size=batch_size,
                          shuffle=True,
                          num_workers=0)

device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
model = NeuralNet(input_size, hidden_size, output_size).to(device)
# Loss and optimizer
criterion = nn.CrossEntropyLoss()
optimizer = torch.optim.Adam(model.parameters(), lr=learning_rate)
# Train the model
for epoch in range(num_epochs):
    for (words, labels) in train_loader:
        words = words.to(device)
        labels = labels.to(dtype=torch.long).to(device)        
        # Forward pass
        outputs = model(words)
        # if y would be one-hot, we must apply
        # labels = torch.max(labels, 1)[1]
        loss = criterion(outputs, labels)        
        # Backward and optimize
        optimizer.zero_grad()
        loss.backward()
        optimizer.step()        
    if (epoch+1) % 100 == 0:
        print (f'Epoch [{epoch+1}/{num_epochs}], Loss: {loss.item():.4f}')

print(f'final loss: {loss.item():.4f}')
data = {
"model_state": model.state_dict(),
"input_size": input_size,
"hidden_size": hidden_size,
"output_size": output_size,
"all_words": all_words,
"tags": tags
}
FILE = "chatdata.pth"
torch.save(data, FILE)
print(f'training complete. file saved to {FILE}')

3.4 - Execute o arquivo train.py anterior para treinar o modelo

O arquivo train.py criará o arquivo chatdata.pth que se utilizará em nosso arquivo chat.py

Passo 4: Criar uma aplicação ChatBot baseada no modelo treinado

Como nosso modelo é criado. No passo final, vamos criar um arquivo chat.py que podemos utilizar no nosso chatbot

#chat.py
import os,random,json,torch
from model import NeuralNet
from nltk_utils import bag_of_words, tokenize

device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')

with open("intents.json") as json_data:
    intents = json.load(json_data)

data_dir = os.path.join(os.path.dirname(__file__))
FILE = os.path.join(data_dir, 'chatdata.pth')
data = torch.load(FILE)

input_size = data["input_size"]
hidden_size = data["hidden_size"]
output_size = data["output_size"]
all_words = data['all_words']
tags = data['tags']
model_state = data["model_state"]

model = NeuralNet(input_size, hidden_size, output_size).to(device)
model.load_state_dict(model_state)
model.eval()

bot_name = "iris-NLP"
def get_response(msg):
    sentence = tokenize(msg)
    X = bag_of_words(sentence, all_words)
    X = X.reshape(1, X.shape[0])
    X = torch.from_numpy(X).to(device)

    output = model(X)
    _, predicted = torch.max(output, dim=1)

    tag = tags[predicted.item()]

    probs = torch.softmax(output, dim=1)
    prob = probs[0][predicted.item()]
    if prob.item() > 0.75:
        for intent in intents['intents']:
            if tag == intent["tag"]:
                return random.choice(intent['responses'])
    return "I do not understand..."
if __name__ == "__main__":
    print("Let's chat! (type 'quit' to exit)")
    while True:
        sentence = input("You: ")
        if sentence == "quit":
            break
        resp = get_response(sentence)
        print(resp)

Ejecutad el archivo chat.py y realizad algunas preguntas que ya definimos anteriormente en nuestro archivo intents.json

 

Para mais informação, visite a página do open exchange IRIS-GenLab

Obrigado