Question

我试图使用InceptionV4来解决某些问题。在使用它之前，我试图尝试使用它。

我用新的密集层替换了最后一个密集层（大小为1001），编译了模型并试图使其适合

from keras import backend as K
import inception_v4
import numpy as np
import cv2
import os

from keras import optimizers
from keras.preprocessing.image import ImageDataGenerator
from keras.models import Sequential
from keras.layers import Convolution2D, MaxPooling2D, ZeroPadding2D
from keras.layers import Activation, Dropout, Flatten, Dense, Input

from keras.models import Model
os.environ['CUDA_VISIBLE_DEVICES'] = ''


my_batch_size=32


train_data_dir ='//shared_directory/projects/try_CDFxx/data/train/'
validation_data_dir ='//shared_directory/projects/try_CDFxx/data/validation/'


img_width, img_height = 299, 299
num_classes=3
nb_epoch=50
nbr_train_samples = 24
nbr_validation_samples = 12


def train_top_model (num_classes):

    v4 = inception_v4.create_model(weights='imagenet')
    predictions = Dense(output_dim=num_classes, activation='softmax', name="newDense")(v4.layers[-2].output) # replacing the 1001 categories dense layer with my own 
    main_input= v4.layers[1].input
    main_output=predictions
    t_model = Model(input=[main_input], output=[main_output])
    train_datagen = ImageDataGenerator(
            rescale=1./255,
            shear_range=0.1,
            zoom_range=0.1,
            rotation_range=10.,
            width_shift_range=0.1,
            height_shift_range=0.1,
            horizontal_flip=True)

    val_datagen = ImageDataGenerator(rescale=1./255)

    train_generator = train_datagen.flow_from_directory(
            train_data_dir,
            target_size = (img_width, img_height),
            batch_size = my_batch_size,
            shuffle = True,
            class_mode = 'categorical')

    validation_generator = val_datagen.flow_from_directory(
            validation_data_dir,
            target_size=(img_width, img_height),
            batch_size=my_batch_size,
            shuffle = True,
            class_mode = 'categorical')
#

    t_model.compile(optimizer='rmsprop', loss='sparse_categorical_crossentropy', metrics=['accuracy'])

#
    t_model.fit_generator(
            train_generator,
            samples_per_epoch = nbr_train_samples,
            nb_epoch = nb_epoch,
            validation_data = validation_generator,
            nb_val_samples = nbr_validation_samples)



train_top_model(num_classes)

但是我收到以下错误

Traceback (most recent call last):
  File "re_try.py", line 76, in <module>
    train_top_model(num_classes)
  File "re_try.py", line 72, in train_top_model
    nb_val_samples = nbr_validation_samples)
  File "/usr/local/lib/python2.7/dist-packages/keras/engine/training.py", line 1508, in fit_generator
    class_weight=class_weight)
  File "/usr/local/lib/python2.7/dist-packages/keras/engine/training.py", line 1261, in train_on_batch
    check_batch_dim=True)
  File "/usr/local/lib/python2.7/dist-packages/keras/engine/training.py", line 985, in _standardize_user_data
    exception_prefix='model target')
  File "/usr/local/lib/python2.7/dist-packages/keras/engine/training.py", line 113, in standardize_input_data
    str(array.shape))
ValueError: Error when checking model target: expected newDense to have shape (None, 1) but got array with shape (24, 3)
Exception in thread Thread-1:
Traceback (most recent call last):
  File "/usr/lib/python2.7/threading.py", line 801, in __bootstrap_inner
    self.run()
  File "/usr/lib/python2.7/threading.py", line 754, in run
    self.__target(*self.__args, **self.__kwargs)
  File "/usr/local/lib/python2.7/dist-packages/keras/engine/training.py", line 409, in data_generator_task
    generator_output = next(generator)
  File "/usr/local/lib/python2.7/dist-packages/keras/preprocessing/image.py", line 693, in next
    x = self.image_data_generator.random_transform(x)
  File "/usr/local/lib/python2.7/dist-packages/keras/preprocessing/image.py", line 403, in random_transform
    fill_mode=self.fill_mode, cval=self.cval)
  File "/usr/local/lib/python2.7/dist-packages/keras/preprocessing/image.py", line 109, in apply_transform
    final_offset, order=0, mode=fill_mode, cval=cval) for x_channel in x]
AttributeError: 'NoneType' object has no attribute 'interpolation'

我做错了什么？为什么在我将newDense图层定义为大小为3后，它应该具有(None,1)形状？

非常感谢

PS 我正在添加模型摘要的结尾

merge_25 (Merge)                 (None, 8, 8, 1536)    0           activation_140[0][0]
                                                                   merge_23[0][0]
                                                                   merge_24[0][0]
                                                                   activation_149[0][0]
____________________________________________________________________________________________________
averagepooling2d_15 (AveragePool (None, 1, 1, 1536)    0           merge_25[0][0]
____________________________________________________________________________________________________
dropout_1 (Dropout)              (None, 1, 1, 1536)    0           averagepooling2d_15[0][0]
____________________________________________________________________________________________________
flatten_1 (Flatten)              (None, 1536)          0           dropout_1[0][0]
____________________________________________________________________________________________________
newDense (Dense)                 (None, 3)             4611        flatten_1[0][0]
====================================================================================================
Total params: 41,210,595
Trainable params: 41,147,427
Non-trainable params: 63,168

Answer 1

好的，问题在于

validation_generator = val_datagen.flow_from_directory(...
        class_mode = 'categorical')

Categorical使您的生成器返回单热编码向量。在您的情况下，3-d一个。但是，您将loss设置为sparse_categorical_crossentropy，并将int作为标签。您应该更改class_mode="sparse"或loss="categorical_crossentropy"。

转移学习失败，因为预期密集层具有形状（无，1）

1 个答案: