python - ValueError : Error when checking input: expected conv2d_1_input to have shape (224, 224, 1) 但得到了形状为 (224, 224, 8) 的数组

Question

我有一个关于签名识别的 CNN 项目，但我得到了标题这样的错误。这是代码

#import lib
import numpy as np
import cv2
import matplotlib.pyplot as plt
import os
import time
#tensorflow lib
import tensorflow
from tensorflow import keras
from keras.models import Sequential
from keras.layers import Dense, Activation, Dropout, Flatten
from keras.layers import Conv2D, MaxPooling2D
from keras.layers.normalization import BatchNormalization
from keras.callbacks import Callback
from keras import backend as K
from keras import optimizers
#sklearn lib
from sklearn.model_selection import train_test_split
from sklearn.metrics import confusion_matrix, precision_score, recall_score, f1_score
from sklearn.metrics import classification_report

#data specification
DIRECTORY = 'C:/Users/MSI GF/Pictures/DataLatih/'
CATEGORIES = ["akMundur", "akTajam", "akLembut", "caMenaik", "caMenurun", "cangkang", "coretanTengah", "garisBawah", "others"]
DATASET = []
IMG_ROWS, IMG_COLS = 224, 224
num_classes = 8
#Load DATASET and create DATASET *once exec
def create_training_data():
    for category in CATEGORIES:
        path = os.path.join(DIRECTORY,category)
        class_num = CATEGORIES.index(category)
        for img in os.listdir(path):
            try:
                #read img and preprocess
                #RGB to grayscale
                img_array = cv2.imread(os.path.join(path,img),cv2.IMREAD_GRAYSCALE)
                #grayscale to threshold
                retval,img_array = cv2.threshold(img_array, 128, 1, cv2.THRESH_BINARY)
                DATASET.append([img_array,class_num])
            except Exception as e:
                pass
        print("Jumlah data: ", len(DATASET))
            
create_training_data()

print("Persiapan Data")
#split training set and test set
X = [] #features
Y = [] #labels

for features, label in DATASET:
    X.append(features)
    Y.append(label)

X = np.array(X).reshape(-1, IMG_ROWS, IMG_COLS, 1)
print("Ukuran DATASET : ", X.shape)
#split X, Y to train and test set
x_train,x_test,y_train,y_test = train_test_split(X, Y, train_size=0.4)
x_train = x_train.astype('float32')
x_test = x_test.astype('float32')
x_train/=255
x_test/=255

print("Ukuran x_train : ", x_train.shape)
print("Ukuran x_test : ", x_test.shape)
print("Ukuran y_train : ", len(y_train))
print("Ukuran y_test : ", len(y_test))

#checking image
#change dimension to plt
print("Contoh lima sampel data x_train")
x_train = np.array(x_train).reshape(-1, IMG_ROWS, IMG_COLS)
plt.figure(figsize=(10,10))
for i in range(10):
    plt.imshow(x_train[i], cmap=plt.cm.gray)
    plt.subplot(5,5,i+1)
    plt.xticks([])
    plt.yticks([])
    plt.grid(False)
    plt.xlabel(CATEGORIES[y_train[i]])
plt.show()

#reshape back to use in learning
x_train = np.array(x_train).reshape(-1, IMG_ROWS, IMG_COLS, 1)
#convert class vector to binary class metrics
y_train = keras.utils.to_categorical(y_train, num_classes)
x_test = keras.utils.to_categorical(x_test, num_classes)

print("Contoh kelas : ")
print(y_train[0])
print(y_test[0])

#begin model, using ALexNet architecture
model = Sequential()
#1st Conv layer
model.add(Conv2D(filters=96, input_shape=[224,224,1], kernel_size=(11,11), strides=(4,4), padding='valid', activation='relu'))
#Max Pooling layer 1
model.add(MaxPooling2D(pool_size=(2,2), strides=(2,2), padding='valid'))

#2nd Conv Layer
model.add(Conv2D(filters=256, kernel_size=(11,11), strides=(1,1), padding='valid', activation='relu'))
#Max Pooling layer 2
model.add(MaxPooling2D(pool_size=(2,2), strides=(2,2), padding='valid'))

#3rd Conv Layer
model.add(Conv2D(filters=384, kernel_size=(3,3), strides=(1,1), padding='valid', activation='relu'))

#4th Conv Layer
model.add(Conv2D(filters=384, kernel_size=(3,3), strides=(1,1), padding='valid', activation='relu'))

#5th Conv Layer
model.add(Conv2D(filters=256, kernel_size=(3,3), strides=(1,1), padding='valid', activation='relu'))
#Max Pooling Layer 3
model.add(MaxPooling2D(pool_size=(2,2), strides=(2,2), padding='valid'))

#passing to Fully Connected Layer
model.add(Flatten())
#1st FC Layer
model.add(Dense(4096, input_shape=(224*224*1,)))
model.add(Activation('relu'))
model.add(Dropout(0.4))

#2nd FC Layer
model.add(Dense(4096))
model.add(Activation('relu'))
model.add(Dropout(0.4))

#3rd FC Layer
model.add(Dense(1000))
model.add(Activation('relu'))
model.add(Dropout(0.4))

#Output layer
model.add(Dense(8))
model.add(Activation('softmax'))

model.summary()

sgd = optimizers.SGD(lr=0.01, momentum=0.9, decay=0.0, nesterov=False)
model.compile(loss='categorical_crossentropy', optimizer=sgd, metrics=['categorical_accuracy'])

#training model
print("Training model")
start_time = time.time()
history = model.fit(x_train, y_train, epochs=100, validation_data=(x_test, y_test))

print("\nTraining Model Selesai")
print("Lama waktu learning: ", (time.time() - start_time) / 60)

print(history.history.keys())
plt.figure(1)

#calculate loss and accuracy
score = model.evaluate(x_test, y_test)
print('Model telah selesai dilakukan pelatihan')
print('Test Loss : ', score[0])
print('Test Accuracy : ', score[1]*100.0)

#save model and weight
#model to json files
model_json = model.to.json()
with open("model/model_json_sgd001.json", "w") as json_file:
    json_file.write(model_json)
#weight to h5 file
model.save_weights("model/model_1_sgd001.h5")

print("Model dan Bobot telah disimpan")

训练模型部分显示的错误。这是错误

--------------------------------------------------------------------------- ValueError                                Traceback (most recent call last) <ipython-input-12-c5d6ddb58171> in <module>
      2 print("Training model")
      3 start_time = time.time()
----> 4 history = model.fit(x_train, y_train, epochs=100, validation_data=(x_test, y_test))
      5 
      6 print("\nTraining Model Selesai")

C:\Anaconda3\envs\PythonGPU\lib\site-packages\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,
**kwargs)
    970                 val_x, val_y,
    971                 sample_weight=val_sample_weight,
--> 972                 batch_size=batch_size)
    973             if self._uses_dynamic_learning_phase():
    974                 val_ins = val_x + val_y + val_sample_weights + [0.]

C:\Anaconda3\envs\PythonGPU\lib\site-packages\keras\engine\training.py in _standardize_user_data(self, x, y, sample_weight, class_weight, check_array_lengths, batch_size)
    749             feed_input_shapes,
    750             check_batch_axis=False,  # Don't enforce the batch size.
--> 751             exception_prefix='input')
    752 
    753         if y is not None:

C:\Anaconda3\envs\PythonGPU\lib\site-packages\keras\engine\training_utils.py in standardize_input_data(data, names, shapes, check_batch_axis, exception_prefix)
    136                             ': expected ' + names[i] + ' to have shape ' +
    137                             str(shape) + ' but got array with shape ' +
--> 138                             str(data_shape))
    139     return data
    140 

ValueError: Error when checking input: expected conv2d_1_input to have shape (224, 224, 1) but got array with shape (224, 224, 8)

我认为这是因为我的输入有 8 个 channel 而不是 1 个，但是如何修复它？

Answer 1

似乎您对输入张量而不是标签进行了一次性编码:

x_test = keras.utils.to_categorical(x_test, num_classes)

我觉得应该是 y_test