Tensorflow 자격증 대비 - 셰익스피어 따라하기 자연어 처리 (NLP with LSTM)

Tensorflow 자격증 

 

1. 관련 라이브러리 

 

from tensorflow.keras.preprocessing.sequence import pad_sequences
from tensorflow.keras.layers import Embedding, LSTM, Dense, Dropout, Bidirectional
from tensorflow.keras.preprocessing.text import Tokenizer
from tensorflow.keras.models import Sequential
from tensorflow.keras.optimizers import Adam
### YOUR CODE HERE
from tensorflow.keras import regularizers
# Figure out how to import regularizers
import tensorflow.keras.utils as ku 
import numpy as np

 

2. 관련 데이터 받기

 

tokenizer = Tokenizer()

# sonnets.txt
!gdown --id 108jAePKK4R3BVYBbYJZ32JWUwxeMg20K

data = open('./sonnets.txt').read()

corpus = data.lower().split("\n")


tokenizer.fit_on_texts(corpus)
total_words = len(tokenizer.word_index) + 1

# create input sequences using list of tokens
input_sequences = []
for line in corpus:
	token_list = tokenizer.texts_to_sequences([line])[0]
	for i in range(1, len(token_list)):
		n_gram_sequence = token_list[:i+1]
		input_sequences.append(n_gram_sequence)


# pad sequences 
max_sequence_len = max([len(x) for x in input_sequences])
input_sequences = np.array(pad_sequences(input_sequences, maxlen=max_sequence_len, padding='pre'))

# create predictors and label
predictors, label = input_sequences[:,:-1],input_sequences[:,-1]

label = ku.to_categorical(label, num_classes=total_words)

 

3. 모델 구성

 

### START CODE HERE
model = Sequential()
model.add(Embedding(total_words, 100, input_length=max_sequence_len-1))# Your Embedding Layer)
model.add(Bidirectional(LSTM(150, return_sequences = True)))# An LSTM Layer)
model.add(Dropout(0.2))# A dropout layer)
model.add(LSTM(100))# Another LSTM Layer)
model.add(Dense(total_words/2, activation='relu', kernel_regularizer=regularizers.l2(0.01)))# A Dense Layer including regularizers)
model.add(Dense(total_words, activation='softmax'))# A Dense Layer)
# Pick an optimizer
model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])# Pick a loss function and an optimizer)
### END CODE HERE

print(model.summary())

 

4. function 을 plot 해봅시다!

 

import matplotlib.pyplot as plt
acc = history.history['accuracy']
loss = history.history['loss']

epochs = range(len(acc))

plt.plot(epochs, acc, 'b', label='Training accuracy')
plt.title('Training accuracy')

plt.figure()

plt.plot(epochs, loss, 'b', label='Training Loss')
plt.title('Training loss')
plt.legend()

plt.show()

5.  실제로 해 봅시다

 

seed_text = "Help me Obi Wan Kenobi, you're my only hope"
next_words = 100
  
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_classes(token_list, verbose=0)
	output_word = ""
	for word, index in tokenizer.word_index.items():
		if index == predicted:
			output_word = word
			break
	seed_text += " " + output_word
print(seed_text)