악플 검사 AI 만들기 2 마지막 모델 구현 답안 없을까요 ㅜㅜ

[코앱]

계속 오류나서 좀 보면서 하고 싶습니당... ㅜㅜ

[codingapple]

만들어놓은건 없습니다

[코앱]

넹 ㅠㅠ 열심히해볼게요..!

[오형선]

부족하지만 제가한거 올립니다

import tensorflow as tf
import numpy as np
from sklearn.model_selection import train_test_split
from tensorflow.keras.preprocessing.text import Tokenizer
from tensorflow.keras.preprocessing.sequence import pad_sequences

raw['review'] = raw['review'].str.replace('[^ㄱ-ㅎㅏ-ㅣ가-힣0-9 ]', '', regex=True)
raw.drop_duplicates(subset=['review'], inplace=True)

unique_review = raw['review'].tolist()
unique_review = ''.join(unique_review)

tokenizer = Tokenizer(char_level=True, oov_token='<OOV>')
unique_review = raw['review'].tolist()
tokenizer.fit_on_texts(unique_review)

train_seq = tokenizer.texts_to_sequences(unique_review)

Y = raw['label'].tolist()
raw['length'] = raw['review'].str.len()

X = pad_sequences(train_seq, maxlen=100, padding='pre')
trainX, valX, trainY, valY = train_test_split(X, Y, test_size=0.2, random_state=42)

trainY = np.array(trainY, dtype=np.float32)
valY = np.array(valY, dtype=np.float32)

model = tf.keras.models.Sequential([
    tf.keras.layers.Input(shape=(100,)),
    tf.keras.layers.Embedding(len(tokenizer.word_index) + 1, 16),
    tf.keras.layers.Bidirectional(tf.keras.layers.LSTM(128)),
    tf.keras.layers.Dropout(0.2),
    tf.keras.layers.Dense(1, activation='sigmoid')
])

model.compile(
    loss='binary_crossentropy',
    optimizer='adam',
    metrics=['accuracy']
)

model.fit(
    trainX, trainY,
    epochs=5,
    batch_size=64,
    validation_data=(valX, valY)
)

model.summary()

[글쓴이]

글