使用keras+tf训练神经网络莫名退出程序

使用keras + tf训练神经网络时出现错误，没有报错直接退出程序

import keras

from keras.optimizers import adam_v2
from keras.models import Sequential
from keras.layers import Dense
from keras.layers import Convolution2D
from keras.layers import Flatten
from keras.layers import MaxPooling2D
from keras.preprocessing.image import ImageDataGenerator
from keras.models import Sequential
from keras.layers import Conv2D
from keras.layers import MaxPooling2D
from keras.layers import Flatten
from keras.layers import Dense
from keras.layers import Input, Conv2D, Dense, concatenate,MaxPooling2D,Dropout
from keras.models import Model
from tensorflow.keras.callbacks import ModelCheckpoint, EarlyStopping
from collections import Counter
from keras.callbacks import EarlyStopping
import tensorflow as tf

train_loc ='D:/Lung_sound_classification-main/Feature_extraction/picture_for_cnn/train/'
validation_loc = 'D:/Lung_sound_classification-main/Feature_extraction/picture_for_cnn/validation/'
trdata = ImageDataGenerator(rescale=1 / 255.0)

traindata = trdata.flow_from_directory(directory=train_loc, target_size=(224,224),batch_size=5,shuffle=False)

valdata = ImageDataGenerator(rescale=1 / 255.0)
validationdata = valdata.flow_from_directory(directory=validation_loc, target_size=(224,224),batch_size=5,shuffle=False)

img_inputs = Input(shape=(224,224, 3))

classifier=Conv2D(64, (5, 5), activation = 'relu')(img_inputs)

classifier=MaxPooling2D(pool_size = (2, 2))(classifier)

classifier=Conv2D(64, (3, 3), activation = 'relu')(classifier)

classifier=MaxPooling2D(pool_size = (2, 2))(classifier)

classifier=Conv2D(96, (3, 3), activation = 'relu')(classifier)

classifier=MaxPooling2D(pool_size = (2, 2))(classifier)

classifier=Conv2D(96, (3, 3), activation = 'relu')(classifier)

classifier=MaxPooling2D(pool_size = (2, 2))(classifier)

classifier=Flatten()(classifier)

classifier=Dense(units = 256, activation = 'relu')(classifier)
classifier=Dropout(0.6)(classifier)
classifier=Dense(units = 128, activation = 'relu')(classifier)
classifier=Dropout(0.3)(classifier)
classifier=Dense(units = 64, activation = 'relu')(classifier)
classifier=Dropout(0.15)(classifier)
classifier=Dense(units = 32, activation = 'relu')(classifier)
classifier=Dropout(0.075)(classifier)
classifier=Dense(units = 16, activation = 'relu')(classifier)
classifier=Dropout(0.0325)(classifier)
classifier=Dense(units = 8, activation = 'relu')(classifier)
outputs=Dense(units = 5, activation = 'softmax')(classifier)

model = Model(inputs=img_inputs, outputs=outputs, name="LS_model")

opt = adam_v2.Adam(learning_rate=0.00001)
print(opt)
model.compile(optimizer = opt, loss = 'categorical_crossentropy', metrics = ['accuracy'])

STEP_SIZE_TRAIN=traindata.n//traindata.batch_size

STEP_SIZE_VALID=validationdata.n//validationdata.batch_size

checkpoint = ModelCheckpoint("working/best_model.h5", monitor='val_accuracy', verbose=1,
save_best_only=True, save_weights_only=False, mode='auto')

early = EarlyStopping(monitor='val_accuracy', min_delta=0, patience=20, verbose=1, mode='auto')
model.summary()

histore = model.fit(traindata,epochs=600,steps_per_epoch=STEP_SIZE_TRAIN,validation_data=validationdata,validation_steps=STEP_SIZE_VALID,callbacks=[checkpoint,early])

我的解答思路和尝试过的方法

有没有哥们回答一下啊！急死啦

主要原因，训练使用内存或显存不足：
1、修改：Batch大小。
2、修改每个加载数据大小
3、减少训练网络复杂程度，减少训练网络涉及到得参数数量，减少训练网络的步骤等。