Question

基本上，我是从udemy那里学习机器学习A-Z教程的，在那里我学会了训练自己的卷积神经网络模型，但是并未向我展示如何使用训练后的模型-如何给出输入并获得输出。我在youtube上查看了一些教程，以通过Keras中的cnn传递单个图像，并附带了this教程。当我运行代码时，出现错误

ValueError：检查输入时出错：预期conv2d_3_input具有形状（64，64，3），但数组具有形状（64，64，1）

现在我完全不知道如何处理这个问题。如果有人可以告诉我在ive培训了我的网络之后该怎么办以及如何解决整个问题，那么请衷心地感谢ive尝试更改img_array = cv2.imread(filepath, cv2.IMREAD_GRAYSCALE) 读取颜色，但我认为与此无关

我用来训练卷积网络的代码：

import keras
from keras.models import Sequential
from keras.layers import Convolution2D
from keras.layers import MaxPooling2D
from keras.layers import Flatten
from keras.layers import Dense

初始化cnn classifier = Sequential()

第1步-应用卷积

classifier.add(Convolution2D(32,3,3, border_mode = 'same', input_shape=(64,64,3), activation = 'relu'))

应用最大池化

classifier.add(MaxPooling2D(pool_size=(2,2)))

添加另一个卷积层以提高准确性

classifier.add(Convolution2D(32,3,3, border_mode = 'same', activation = 'relu'))
classifier.add(MaxPooling2D(pool_size=(2,2)))

应用展平

classifier.add(Flatten())

第4步-完全连接

classifier.add(Dense(output_dim = 128, activation = 'relu'))
#input
classifier.add(Dense(output_dim = 1, activation = 'sigmoid'))
#output

全部编译

classifier.compile(optimizer = 'adam', loss = 'binary_crossentropy', metrics = ['accuracy'])
#fitting the cnn to the images
from keras.preprocessing.image import ImageDataGenerator
#pixels take values between 0 & 255
train_datagen = ImageDataGenerator(rescale=1./255,                        
    shear_range=0.2,
    zoom_range=0.2,
    horizontal_flip=True)
test_datagen = ImageDataGenerator(rescale=1./255)

training_set = train_datagen.flow_from_directory('dataset/training_set', target_size=(64, 64), batch_size=32, class_mode='binary')       
test_set = test_datagen.flow_from_directory('dataset/test_set',
                                         target_size=(64, 64),
                                         batch_size=32,
                                         class_mode='binary')

classifier.fit_generator(training_set,
                     samples_per_epoch = 8000,
                     nb_epoch = 25,
                     verbose = 1,
                     validation_data = test_set,
                     nb_val_samples = 2000)

classifier.save('my_model.h5')

我在新文件中使用的代码来尝试训练有素的网络

import cv2
import tensorflow as tf
import keras
from keras.models import Sequential

Categories = ["Dogs","cats"]

def prepare(filepath):
    img_size = 64
    img_array = cv2.imread(filepath, cv2.IMREAD_GRAYSCALE)
    new_array = cv2.resize(img_array, (img_size,img_size))
    return new_array.reshape(-1,img_size,img_size, 1)
    classifier = keras.models.load_model('my_model.h5')

prediction = classifier.predict([prepare('/Users/m.zain/Documents/machine learning-2019 dps/Machine Learning A-Z Template Folder/Part 8 - Deep Learning/Section 40 - Convolutional Neural Networks (CNN)/dog.jpg')])
print(prediction)

prediction = classifier.predict([prepare('/Users/m.zain/Documents/machine learning-2019 dps/Machine Learning A-Z Template Folder/Part 8 - Deep Learning/Section 40 - Convolutional Neural Networks (CNN)/dog.jpg')])
 print(Categories[int(prediction[0][0])])

prediction = classifier.predict([prepare('/Users/m.zain/Documents/machine learning-2019 dps/Machine Learning A-Z Template Folder/Part 8 - Deep Learning/Section 40 - Convolutional Neural Networks (CNN)/cat.jpg')])
 print(Categories[int(prediction[0][0])])

我希望在执行代码时得到“ dog”作为输出：

prediction = classifier.predict([prepare('/Users/m.zain/Documents/machine learning-2019 dps/Machine Learning A-Z Template Folder/Part 8 - Deep Learning/Section 40 - Convolutional Neural Networks (CNN)/dog.jpg')])
print(Categories[int(prediction[0][0])])

当我运行它时将cat作为输出：

prediction = classifier.predict([prepare('/Users/m.zain/Documents/machine learning-2019 dps/Machine Learning A-Z Template Folder/Part 8 - Deep Learning/Section 40 - Convolutional Neural Networks (CNN)/cat.jpg')])
print(Categories[int(prediction[0][0])])

检查输入时出错：预期conv2d_3_input具有形状（64、64、3），但数组形状为（64、64、1）

0 个答案: