检查目标时出错:预期density_3的形状为(3，)，但数组的形状为(1，) [英] Error when checking target: expected dense_3 to have shape (3,) but got array with shape (1,)

问题描述

我正在Places205的3类子集上的Keras中训练VGG16样模型，并遇到以下错误:

I am working on training a VGG16-like model in Keras, on a 3 classes subset from Places205, and encountered the following error:

ValueError: Error when checking target: expected dense_3 to have shape (3,) but got array with shape (1,)

我阅读了多个类似的问题，但到目前为止，没有一个问题对我有帮助.错误出现在最后一层，我放置了3个，因为这是我现在正在尝试的类的数量.

I read multiple similar issues but none helped me so far. The error is on the last layer, where I've put 3 because this is the number of classes I'm trying right now.

代码如下:

import keras from keras.datasets
import cifar10 from keras.preprocessing.image 
import ImageDataGenerator from keras.models 
import Sequential 
from keras.layers import Dense, Dropout, Activation, Flatten from keras.layers import Conv2D, MaxPooling2D 
from keras import backend as K import os


# Constants used  
img_width, img_height = 224, 224  
train_data_dir='places\\train'  
validation_data_dir='places\\validation'  
save_filename = 'vgg_trained_model.h5'  
training_samples = 15  
validation_samples = 5  
batch_size = 5  
epochs = 5


if K.image_data_format() == 'channels_first':
    input_shape = (3, img_width, img_height) else:
    input_shape = (img_width, img_height, 3)

model = Sequential([
    # Block 1
    Conv2D(64, (3, 3), activation='relu', input_shape=input_shape, padding='same'),
    Conv2D(64, (3, 3), activation='relu', padding='same'),
    MaxPooling2D(pool_size=(2, 2), strides=(2, 2)),
    # Block 2
    Conv2D(128, (3, 3), activation='relu', padding='same'),
    Conv2D(128, (3, 3), activation='relu', padding='same'),
    MaxPooling2D(pool_size=(2, 2), strides=(2, 2)),
    # Block 3
    Conv2D(256, (3, 3), activation='relu', padding='same'),
    Conv2D(256, (3, 3), activation='relu', padding='same'),
    Conv2D(256, (3, 3), activation='relu', padding='same'),
    MaxPooling2D(pool_size=(2, 2), strides=(2, 2)),
    # Block 4
    Conv2D(512, (3, 3), activation='relu', padding='same'),
    Conv2D(512, (3, 3), activation='relu', padding='same'),
    Conv2D(512, (3, 3), activation='relu', padding='same'),
    MaxPooling2D(pool_size=(2, 2), strides=(2, 2)),
    # Block 5
    Conv2D(512, (3, 3), activation='relu', padding='same',),
    Conv2D(512, (3, 3), activation='relu', padding='same',),
    Conv2D(512, (3, 3), activation='relu', padding='same',),
    MaxPooling2D(pool_size=(2, 2), strides=(2, 2)),
    # Top
    Flatten(),
    Dense(4096, activation='relu'),
    Dense(4096, activation='relu'),
    Dense(3, activation='softmax') ])

model.summary()

model.compile(loss='binary_crossentropy',
              optimizer='rmsprop',
              metrics=['accuracy'])

# no augmentation config train_datagen = ImageDataGenerator() validation_datagen = ImageDataGenerator()
     train_generator = train_datagen.flow_from_directory(
    train_data_dir,
    target_size=(img_width, img_height),
    batch_size=batch_size,
    class_mode='binary')

validation_generator = validation_datagen.flow_from_directory(
    validation_data_dir,
    target_size=(img_width, img_height),
    batch_size=batch_size,
    class_mode='binary')

model.fit_generator(
    train_generator,
    steps_per_epoch=training_samples // batch_size,
    epochs=epochs,
    validation_data=validation_generator,
    validation_steps=validation_samples // batch_size)

model.save_weights(save_filename)

推荐答案

问题出在标签数据形状上.在多类别问题中，您正在预测每个可能类别的概率，因此必须提供(N，m)形状的标签数据，其中N是训练示例的数量，m是可能类别的数量(在您的情况下为3)

The problem is with your label-data shape. In a multiclass problem you are predicting the probabibility of every possible class, so must provide label data in (N, m) shape, where N is the number of training examples, and m is the number of possible classes (3 in your case).

Keras期望y数据呈(N，3)形状，而不是您可能提供的(N，)形状，这就是为什么它会引发错误.

Keras expects y-data in (N, 3) shape, not (N,) as you've problably provided, that's why it raises an error.

使用例如 OneHotEncoder 可以将标签数据转换为一键编码表格.

Use e.g. OneHotEncoder to convert your label data to one-hot encoded form.

这篇关于检查目标时出错:预期density_3的形状为(3，)，但数组的形状为(1，)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！