ValueError:此损失期望目标与输出具有相同的形状 [英] ValueError : This loss expects targets to have the same shape as the output
问题描述
我正在尝试使用卷积神经网络对猫和狗进行分类.X包含特征,Y包含标签.图像数据集来自Microsoft网站.我不明白为什么会出现形状错误.
I was trying to classify the Cats and Dogs using Convolutional Neural Network. Here X contains the featuers and Y the labels. The image dataset is taken from Microsoft site. I dont understand why the shape error is coming.
Image_Size=65
import pickle
import numpy as np
X=np.asarray(pickle.load(open("X.pickle","rb")))
Y=np.asarray(pickle.load(open("X.pickle","rb")))
import tensorflow as tf
from tensorflow.keras.layers import Dense, Activation, Flatten,Conv2D, MaxPooling2D
from tensorflow.keras.models import Sequential
X=X/255.0
model=Sequential()
model.add(Conv2D(64,(3,3), input_shape=X.shape[1:]))
model.add(Activation('relu'))
model.add(MaxPooling2D(pool_size=(2,2)))
model.add(Conv2D(64,(3,3)))
model.add(Activation('relu'))
model.add(MaxPooling2D(pool_size=(2,2)))
model.add(Flatten())
model.add(Dense(64))
model.add(Activation('relu'))
model.add(Dense(1))
model.add(Activation('sigmoid'))
model.compile(loss='binary_crossentropy',optimizer='adam',metrics=['accuracy'])
model.fit(X,Y,batch_size=32,epochs=3,validation_split=0.3)
现在,我收到以下错误消息.
Now I get the following error.
---------------------------------------------------------------------------
ValueError Traceback (most recent call last)
<ipython-input-67-1eaf182a2702> in <module>
23
24 model.compile(loss='binary_crossentropy',optimizer='adam',metrics=['accuracy'])
---> 25 model.fit(X,Y,batch_size=32,epochs=3,validation_split=0.3)
26 model.summary()
C:\Anaconda\lib\site-packages\tensorflow_core\python\keras\engine\training.py in fit(self, x, y,
batch_size, epochs, verbose, callbacks, validation_split, validation_data, shuffle, class_weight,
sample_weight, initial_epoch, steps_per_epoch, validation_steps, validation_freq, max_queue_size,
workers, use_multiprocessing, **kwargs)
726 max_queue_size=max_queue_size,
727 workers=workers,
--> 728 use_multiprocessing=use_multiprocessing)
729
730 def evaluate(self,
C:\Anaconda\lib\site-packages\tensorflow_core\python\keras\engine\training_v2.py in fit(self, model,
x, y, batch_size, epochs, verbose, callbacks, validation_split, validation_data, shuffle,
class_weight, sample_weight, initial_epoch, steps_per_epoch, validation_steps, validation_freq,
**kwargs)
222 validation_data=validation_data,
223 validation_steps=validation_steps,
--> 224 distribution_strategy=strategy)
225
226 total_samples = _get_total_number_of_samples(training_data_adapter)
C:\Anaconda\lib\site-packages\tensorflow_core\python\keras\engine\training_v2.py in
_process_training_inputs(model, x, y, batch_size, epochs, sample_weights, class_weights,
steps_per_epoch, validation_split, validation_data, validation_steps, shuffle,
distribution_strategy, max_queue_size, workers, use_multiprocessing)
514 batch_size=batch_size,
515 check_steps=False,
--> 516 steps=steps_per_epoch)
517 (x, y, sample_weights,
518 val_x, val_y,
C:\Anaconda\lib\site-packages\tensorflow_core\python\keras\engine\training.py in
_standardize_user_data(self, x, y, sample_weight, class_weight, batch_size, check_steps,
steps_name,
steps, validation_split, shuffle, extract_tensors_from_dataset)
2536 # Additional checks to avoid users mistakenly using improper loss fns.
2537 training_utils.check_loss_and_target_compatibility(
-> 2538 y, self._feed_loss_fns, feed_output_shapes)
2539
2540 # If sample weight mode has not been set and weights are None for all the
C:\Anaconda\lib\site-packages\tensorflow_core\python\keras\engine\training_utils.py in
check_loss_and_target_compatibility(targets, loss_fns, output_shapes)
741 raise ValueError('A target array with shape ' + str(y.shape) +
742 ' was passed for an output of shape ' + str(shape) +
--> 743 ' while using as loss `' + loss_name + '`. '
744 'This loss expects targets to have the same shape '
745 'as the output.')
ValueError: A target array with shape (24946, 65, 65, 1) was passed for an output of shape (None, 1)
while using as loss `binary_crossentropy`. This loss expects targets to have the same shape as the
output.
如果有人在这件事上提供帮助,那就太好了.
It would be great if someone helps in this matter.
推荐答案
最简单的答案是:当标签为图像时,您的网络仅返回一个数字.如果您要分类图像上的猫还是狗,则标签应为数字.例如,如果图像上是猫,则为 1
,而 0
.特征应该是表示为3维张量的图像.在您的代码中
The short answer is: your network returns just one number when your labels are images.
If you would classify if on the image is a cat or a dog your labels should be a numbers. For example 1
if on the image is a cat and 0
. The features should be images represented as 3 dimensional tensors.
In your code
X=np.asarray(pickle.load(open("X.pickle","rb")))
Y=np.asarray(pickle.load(open("X.pickle","rb")))
X
和 Y
具有相同的对象-您没有标签和特征(您可能具有两倍的特征).
X
and Y
have the same objects - you don't have labels and features (you have probably two times features).
第二个问题可能在这一行:
The second problem could be in this line:
model.fit(X,Y,batch_size=32,epochs=3,validation_split=0.3)
正如您提到的,您在 X
-标签和i Y
功能中.方法 fit
将作为第一个参数特征并作为第二个参数标签.
As you mentioned you have in X
- labels, and i Y
features. The method fit
takes as first argument features and as the second argument labels.
请注意表示法.通常,在深度学习问题中, X
表示功能,而 Y
表示标签.
Please be careful about the notation. Typically in deeplearning problems X
denotes features and Y
denotes labels.
这篇关于ValueError:此损失期望目标与输出具有相同的形状的文章就介绍到这了,希望我们推荐的答案对大家有所帮助,也希望大家多多支持IT屋!