为什么迁移学习和数据增强的准确度得分较低？

Question

我正在开发分类模型。我有 5 种花，每种花大约 1000 张照片。

我有两个问题。

1. 我使用 VGG16 和迁移学习作为模型，准确度得分为 84。为了提高准确度分数，我尝试了数据增强。但是，当我同时使用迁移学习和数据增强时，准确率下降到 %74。

2. 如果不能同时使用迁移学习和数据增强，我可以将增强的照片保存到训练文件夹，然后使用迁移学习吗？

   select Batsman_, sum(batsman_runs)/count(player_dismissed) as Average 
   from
   (
    (select batsman as Batsman_ from IPL_BALL_BY_BALL) 
    union all 
    (select non_striker as Batsman_ from IPL_BALL_BY_BALL)
   )
   group by Batsman_ 
   order by Average desc;
   

   def define_model_vgg_16():
   # load model
   model = VGG16(include_top=False,input_shape=(224,224,3))
   # mark loaded layers as not trainable
   for layer in model.layers:
       layer.trainable = False
   # add new classifier layers
   flat1 = Flatten()(model.layers[-1].output)
   class1 = Dense(128,activation='relu',kernel_initializer='he_uniform')(flat1)
   output = Dense(5,activation='softmax')(class1)
   # define new model
   model = Model(inputs=model.inputs, outputs = output)
   #compile model
   opt = SGD(lr=0.001,momentum=0.9)
   model.compile(optimizer=opt,loss='categorical_crossentropy',metrics=['accuracy'])
   return model