제발 도와주세요ㅜㅠ

import numpy as np
from keras.models import Sequential
from keras.layers import Dense, Dropout, Flatten, LSTM
from keras.layers import BatchNormalization
from keras.initializers import he_normal
!pip install np_utils
import np_utils
from tensorflow.keras.utils import to_categorical
from google.colab import drive

drive.mount('/content/drive')

[train_x, train_y, test_x, test_y] = np.load('/content/drive/MyDrive/LSTM_disaster_prediction_data.npy', allow_pickle = True)

X_train = train_x.astype('float32')
X_test = test_x.astype('float32')

X_train = np.transpose(X_train,(0,2,1))
X_test = np.transpose(X_test,(0,2,1))

y_train = to_categorical(train_y, num_classes=num_classes)
y_test = to_categorical(test_y, num_classes=num_classes)

num_classes = y_train.shape[1]

model = Sequential()
model.add(LSTM(units = 4, return_sequences = True, input_shape = (168, 12)))
model.add(LSTM(units = 1, return_sequences= True))
model.add(BatchNormalization())
model.add(Dense(64,activation = 'relu', kernel_initializer = 'he_normal'))
model.add(Dropout(0.25))
model.add(Dense(num_classes, activation = 'softmax'))

model.summary()

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

model.fit(X_train, y_train, epochs = 5, batch_size = 16, verbose = 1, validation_data=(X_test, y_test))

test_loss, test_accuracy = model.evaluate(X_test, y_test)

print('검증결과')
print('test loss = ', test_loss)
print('test accuracy = ', test_accuracy)

ValueError: Shapes (None, 3) and (None, 168, 3) are incompatible

이렇게 뜨는데 어떻게 고쳐야할지 모르겠어요 ㅠㅠㅠㅠㅠㅠㅠㅠㅠㅠ

...
# num_classes 정의를 to_categorical 사용 전으로 이동
num_classes = np.max(train_y) + 1

y_train = to_categorical(train_y, num_classes=num_classes)
y_test = to_categorical(test_y, num_classes=num_classes)
...

...
model.add(LSTM(units = 4, return_sequences = True, input_shape = (168, 12)))
model.add(LSTM(units = 1, return_sequences = False))  # 여기를 수정
model.add(BatchNormalization())
...