使用python和Keras将多个train_datagen组合在一起
我有多个train_datagen(3),我需要将它们组合在一个中,然后使用生成的组合数据生成器拟合模型,但它不适合我使用python和Keras将多个train_datagen组合在一起,python,keras,neural-network,conv-neural-network,Python,Keras,Neural Network,Conv Neural Network,我有多个train_datagen(3),我需要将它们组合在一个中,然后使用生成的组合数据生成器拟合模型,但它不适合我 train_datagen1 = ImageDataGenerator(rescale=1. / 255,rotation_range=15) train_generator1 = train_datagen1.flow_from_directory( train_data_dir, target_size=(224, 224),
train_datagen1 = ImageDataGenerator(rescale=1. / 255,rotation_range=15)
train_generator1 = train_datagen1.flow_from_directory(
train_data_dir,
target_size=(224, 224),
batch_size=batch_size,
class_mode='categorical', shuffle= True)
train_datagen2 = ImageDataGenerator(rescale=1. / 255,rotation_range=20)
train_generator2 = train_datagen2.flow_from_directory(
train_data_dir,
target_size=(224, 224),
batch_size=batch_size,
class_mode='categorical', shuffle= True)
train_datagen3 = ImageDataGenerator(rescale=1. / 255,rotation_range=30)
train_generator3 = train_datagen3.flow_from_directory(
train_data_dir,
target_size=(224, 224),
batch_size=batch_size,
class_mode='categorical', shuffle= True)
trains_gene_tab=zip(train_generator1, train_generator2,train_generator3)
history= model.fit_generator(
trains_gene_tab,
samples_per_epoch=120,
epochs=epochs,
validation_data=validation_generator,
validation_steps=50
)
你能详细说明一下吗?是否有错误当程序执行历史记录行时,模型在开始时停止拟合,即epoch 1/100,然后显示一个大的numpy数组并完成。