chatbot_model.py

import nltk
from nltk.stem import WordNetLemmatizer
import numpy as np
import json
import pickle
import random
# Tensorflow and Keras for Neural Network (AI Model)
import tensorflow as tf
from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Dense, Dropout # <--- YE LINE ZAROORI HAI
from tensorflow.keras.optimizers import Adam

print("All libraries imported successfully.") # <--- YE LINE ADD KAREIN
# Initialisation
# ... (rest of the initialization code) ...

# Initialisation
lemmatizer = WordNetLemmatizer()
words = []
classes = []
documents = []
ignore_words = ['?', '!', '.', ',']


# 1. Load Data
try:
    with open('intents.json', encoding='utf-8') as file:
        data = json.load(file)
    print("Intents file loaded successfully.") 
except Exception as e:
    # Yeh 'Exception as e' sabhi tarah ke errors ko pakadta hai (JSON, File Not Found, etc.)
    print("\n--- CRITICAL ERROR ---")
    print(f"Chatbot failed to start due to: {e}")
    print("Please check intents.json format and location.")
    print("------------------------\n")
    exit() # Agar error aaye to code yahaan ruk jaayega

# 2. Pre-processing
for intent in data['intents']:
    for pattern in intent['patterns']:
        # Tokenize (split into words)
        w = nltk.word_tokenize(pattern)
        words.extend(w)
        # Add to documents list
        documents.append((w, intent['tag']))
        # Add to classes list
        if intent['tag'] not in classes:
            classes.append(intent['tag'])

# Lemmatize (reduce word to its root) and clean up
words = [lemmatizer.lemmatize(w.lower()) for w in words if w not in ignore_words]
words = sorted(list(set(words)))
classes = sorted(list(set(classes)))

print(f"{len(documents)} documents (patterns) loaded")
print(f"{len(classes)} classes (tags) loaded: {classes}")

# 3. Create Training Data (Bag of Words)
training = []
output_empty = [0] * len(classes)

for doc in documents:
    bag = []
    pattern_words = doc[0]
    pattern_words = [lemmatizer.lemmatize(word.lower()) for word in pattern_words]
    
    # Create the Bag of Words (BoW) array
    for w in words:
        bag.append(1) if w in pattern_words else bag.append(0)

    # Output row: one-hot encoding for the intent tag
    output_row = list(output_empty)
    output_row[classes.index(doc[1])] = 1
    
    training.append([bag, output_row])

# Shuffle and convert to NumPy array
random.shuffle(training)
training = np.array(training, dtype=object)

# Create train_x (patterns) and train_y (intents)
train_x = list(training[:,0])
train_y = list(training[:,1])


# 4. Neural Network Model Definition (The AI Brain)
model = Sequential()
# Input layer and first hidden layer
model.add(Dense(128, input_shape=(len(train_x[0]),), activation='relu'))
model.add(Dropout(0.5))
# Second hidden layer
model.add(Dense(64, activation='relu'))
model.add(Dropout(0.5))
# Output layer (predicts the class/tag)
model.add(Dense(len(train_y[0]), activation='softmax'))

# Compile the model (Optimizer helps adjust weights)
adam = Adam(learning_rate=0.01)
model.compile(loss='categorical_crossentropy', optimizer=adam, metrics=['accuracy'])

# 5. Train the Model
print("\n--- Model Training Started ---")
# Adjust epochs/batch_size based on your computer's speed/memory
hist = model.fit(np.array(train_x), np.array(train_y), epochs=200, batch_size=5, verbose=1) 
print("--- Model Training Complete ---")

# 6. Save Model and Data
model.save('chatbot_model.h5',)
pickle.dump({'words':words, 'classes':classes, 'train_x':train_x, 'train_y':train_y}, open( "training_data.pkl", "wb" ))
print("\nModel and data files saved successfully!")

# Model Training Complete hone ke baad:

# Model ko save karein (Sahi tareeka: sirf file ka naam)
model.save('chatbot_model.h5') 

# Data ko save karein
pickle.dump({'words':words, 'classes':classes, 'train_x':train_x, 'train_y':train_y}, open( "training_data.pkl", "wb" ))

# Success message print karein
print("\nModel and data files saved successfully!")