ImageDataGenerator範例問題 - Cupoy

2020/03/13 02:04 PM

電腦視覺深度學習論壇

陽明益

觀看數：0

回答數：3

收藏數：4

cvdl-2

所給的範例中, 這個應該是無法執行的吧~~?!!

就這樣給一個無法執行的範例, 要學員們進行學習, 我覺得教學太跳了

沒有學過的學員, 應該無法吸收的, 應該要檢討一下了~~

#Training Generator

train_datagen = ImageDataGenerator(rescale = 2,shear_range = 0.2,zoom_range = 0.2,horizontal_flip = True)

#Test Generator，只需要Rescale，不需要其他增強

test_datagen = ImageDataGenerator(rescale = 1./255)

#將路徑給Generator，自動產生Label

training_set = train_datagen.flow_from_directory('dataset/training_set',

                                                target_size = (64, 64),

                                                batch_size = 32,

                                                class_mode = 'categorical')

test_set = test_datagen.flow_from_directory('dataset/test_set',

                                           target_size = (64, 64),

                                           batch_size = 32,

                                           class_mode = 'categorical')

#訓練

classifier.fit_generator(training_set,steps_per_epoch = 250,nb_epoch = 25,

                        validation_data = valid_set,validation_steps = 63)