import numpy as np
from tensorflow import keras
from tensorflow.keras.preprocessing.text import Tokenizer
from tensorflow.keras.preprocessing.sequence import pad_sequences

def prepare_and_train_model(file_path, epochs):
    # Считывание данных из файла
    with open(file_path, encoding='utf-8') as f:
        data = f.read()

    # Создание токенизатора
    tokenizer = Tokenizer()
    tokenizer.fit_on_texts([data])

    # Преобразование текста в последовательности чисел
    sequences = tokenizer.texts_to_sequences([data])

    # Создание обучающих данных
    input_sequences = []
    for sequence in sequences:
        for i in range(1, len(sequence)):
            n_gram_sequence = sequence[:i+1]
            input_sequences.append(n_gram_sequence)

    # Предобработка для получения одинаковой длины последовательностей
    max_sequence_len = max([len(sequence) for sequence in input_sequences])
    input_sequences = pad_sequences(input_sequences, maxlen=max_sequence_len, padding='pre')

    # Разделение на входные и выходные данные
    x, y = input_sequences[:, :-1], input_sequences[:, -1]

    # Создание модели рекуррентной нейронной сети
    model = keras.Sequential([
        keras.layers.Embedding(len(tokenizer.word_index) + 1, 100, input_length=max_sequence_len-1),
        keras.layers.Dropout(0.2),
        keras.layers.LSTM(150),
        keras.layers.Dense(len(tokenizer.word_index) + 1, activation='softmax')
    ])

    # Компиляция и обучение модели
    model.compile(loss='sparse_categorical_crossentropy', optimizer='adam', metrics=['accuracy'])
    model.fit(x, y, epochs=epochs, verbose=1)

    return model, tokenizer, max_sequence_len

def generate_text_from_model(model, tokenizer, max_sequence_len, seed_text, next_words):
    # Генерация текста
    for _ in range(next_words):
        token_list = tokenizer.texts_to_sequences([seed_text])[0]
        token_list = pad_sequences([token_list], maxlen=max_sequence_len-1, padding='pre')
        predicted = model.predict(token_list)
        predict_index = np.argmax(predicted, axis=-1)
        word = tokenizer.index_word.get(predict_index[0], '')
        seed_text += " " + word

    return seed_text

model_rus, tokenizer_rus, max_sequence_len_rus = prepare_and_train_model('russian.txt', 150)
rus_text_generated = generate_text_from_model(model_rus, tokenizer_rus, max_sequence_len_rus, "В", 55)

model_eng, tokenizer_eng, max_sequence_len_eng = prepare_and_train_model('english.txt', 150)
eng_text_generated = generate_text_from_model(model_eng, tokenizer_eng, max_sequence_len_eng, "In the", 69)

with open('russian_generated.txt', 'w', encoding='utf-8') as f_rus:
    f_rus.write(rus_text_generated)

with open('english_generated.txt', 'w', encoding='utf-8') as f_eng:
    f_eng.write(eng_text_generated)