【问题标题】:AttributeError: module 'keras.applications.mobilenet' has no attribute 'preprocess'AttributeError:模块“keras.applications.mobilenet”没有属性“预处理”
【发布时间】:2021-02-01 05:11:30
【问题描述】:

我已经导入了 preprocess 模块,但它一直显示相同的错误。我应该如何解决这个错误? 需要帮助来解决这个错误!

导入的库如下所示,

import tensorflow as tf
import keras
from tensorflow.keras.preprocessing import image_dataset_from_directory
from keras.preprocessing.image import ImageDataGenerator
from keras.applications import MobileNet
from keras.applications.mobilenet import preprocess_input

图像准备,

train_path = 'eggplant/training'
test_path = 'eggplant/testing'
valid_path = 'eggplant/validation'

train_batches = ImageDataGenerator(preprocessing_function = keras.applications.mobilenet.preprocess.input).flow_from.directory(train_path, target_size=(224,224), batch_size=10)
train_batches = ImageDataGenerator(preprocessing_function = keras.applications.mobilenet.preprocess.input).flow_from.directory(valid_path, target_size=(224,224), batch_size=10)
train_batches = ImageDataGenerator(preprocessing_function = keras.applications.mobilenet.preprocess.input).flow_from.directory(test_path, target_size=(224,224), batch_size=10, shuffle=False)

错误如下,

AttributeError                            Traceback (most recent call last)
<ipython-input-17-fc384b61e2b7> in <module>()
----> 1 train_batches = ImageDataGenerator(preprocessing_function = keras.applications.mobilenet.preprocess.input).flow_from.directory(train_path, target_size=(224,224), batch_size=10)
      2 train_batches = ImageDataGenerator(preprocessing_function = keras.applications.mobilenet.preprocess.input).flow_from.directory(valid_path, target_size=(224,224), batch_size=10)
      3 train_batches = ImageDataGenerator(preprocessing_function = keras.applications.mobilenet.preprocess.input).flow_from.directory(test_path, target_size=(224,224), batch_size=10, shuffle=False)

AttributeError: module 'keras.applications.mobilenet' has no attribute 'preprocess'

【问题讨论】:

    标签: keras deep-learning transfer-learning image-preprocessing


    【解决方案1】:

    您应该使用preprocess_input 而不是preprocess.input。此外,您还需要将flow_from.directory 更改为flow_from_directory

    【讨论】:

      猜你喜欢
      • 1970-01-01
      • 1970-01-01
      • 2019-03-09
      • 2019-01-14
      • 1970-01-01
      • 2018-01-24
      • 2018-04-14
      • 2019-02-18
      • 1970-01-01
      相关资源
      最近更新 更多