model = models.Sequential()
model.add(layers.Conv2D(32, (3, 3), activation='relu',input_shape=(150, 150,
3)))
model.add(layers.MaxPooling2D((2, 2)))

model.add(layers.Dropout(0.15))
model.add(layers.Conv2D(64, (3, 3), activation='relu'))
model.add(layers.MaxPooling2D((2, 2)))

model.add(layers.Dropout(0.2))
model.add(layers.Conv2D(128, (3, 3), activation='relu'))
model.add(layers.MaxPooling2D((2, 2)))

model.add(layers.Conv2D(256, (3, 3), activation='relu'))
model.add(layers.MaxPooling2D((2, 2)))

model.add(layers.Conv2D(256, (3, 3), activation='relu'))
model.add(layers.MaxPooling2D((2, 2)))
model.add(BatchNormalization())

model.add(layers.Flatten())
model.add(layers.Dropout(0.6))
model.add(layers.Dense(150, activation='relu',
kernel_regularizer=regularizers.l2(0.002)))
model.add(layers.Dense(5, activation='softmax'))

model.compile(loss='categorical_crossentropy',
optimizer=optimizers.Adam(lr=1e-3),
metrics=['acc'])

Epoch 00067: val_loss did not improve from 0.08283
Epoch 68/200
230/230 [==============================] - 56s 243ms/step - loss: 0.0893 -
acc: 0.9793 - val_loss: 0.0876 - val_acc: 0.9784

Epoch 00068: val_loss did not improve from 0.08283
Epoch 69/200
230/230 [==============================] - 58s 250ms/step - loss: 0.0874 -
acc: 0.9774 - val_loss: 0.1209 - val_acc: 0.9684

Epoch 00069: val_loss did not improve from 0.08283
Epoch 70/200
230/230 [==============================] - 57s 246ms/step - loss: 0.0879 -
acc: 0.9803 - val_loss: 0.1384 - val_acc: 0.9706

Epoch 00070: val_loss did not improve from 0.08283
Epoch 71/200
230/230 [==============================] - 59s 257ms/step - loss: 0.0903 -
acc: 0.9783 - val_loss: 0.1352 - val_acc: 0.9728

Epoch 00071: val_loss did not improve from 0.08283
Epoch 72/200
230/230 [==============================] - 58s 250ms/step - loss: 0.0852 -
acc: 0.9798 - val_loss: 0.1324 - val_acc: 0.9621

Epoch 00072: val_loss did not improve from 0.08283
Epoch 73/200
230/230 [==============================] - 58s 250ms/step - loss: 0.0831 -
acc: 0.9815 - val_loss: 0.1634 - val_acc: 0.9574

Epoch 00073: val_loss did not improve from 0.08283
Epoch 74/200
230/230 [==============================] - 57s 246ms/step - loss: 0.0824 -
acc: 0.9816 - val_loss: 0.1280 - val_acc: 0.9640

Epoch 00074: val_loss did not improve from 0.08283
Epoch 75/200
230/230 [==============================] - 57s 247ms/step - loss: 0.0869 -
acc: 0.9774 - val_loss: 0.0777 - val_acc: 0.9882

Epoch 00075: val_loss improved from 0.08283 to 0.07765, saving model to
C:/Users/xxx/Desktop/best_model_7.h5
Epoch 76/200
230/230 [==============================] - 56s 243ms/step - loss: 0.0739 -
acc: 0.9851 - val_loss: 0.0683 - val_acc: 0.9851

Epoch 00076: val_loss improved from 0.07765 to 0.06826, saving model to
C:/Users/xxx/Desktop/best_model_7.h5