如何使用 tensorflow keras 模型识别大于 9 的数字和浮点数？

Question

无论如何使用带有CNN模型的tensorflow keras来训练和识别大于9的数字或十进制数字？我的意思是值域在0 到20 之间，或者从0 到10 之间的小数，比如0,1; 0,2;......;9,9？

我已经用 MNIST 数据集训练了模型，用于从 0 到 9 的数字，但我想理解大于 9 的小数和值域。希望看到你对此的想法......

例如号码10

或号码11,12,....,17,...

更新 我通过上面给定的图像尝试使用带有一些新数字的训练模型，但预测仍然给出错误值。这是一些代码：

def trainDigitsModel(alwaysTrain=False, writeFile=True):
    modelExists = os.path.exists('traineddata/traineddata.model/saved_model.pb')
    
    if modelExists is False or alwaysTrain is True:
        # load data
        mnist = tf.keras.datasets.mnist
        (X_train, y_train), (X_test, y_test) = mnist.load_data()
        
        # Reshaping to format which CNN expects (batch, height, width, channels)
        X_train = X_train.reshape(X_train.shape[0], X_train.shape[1], X_train.shape[2], 1).astype('float32')
        X_test = X_test.reshape(X_test.shape[0], X_test.shape[1], X_test.shape[2], 1).astype('float32')
        
        # Load your own images to training and test data
        X_train, y_train = load_images_to_data('path/to/trainmore', X_train, y_train)
        X_test, y_test = load_images_to_data('path/to/trainmore', X_test, y_test)
        
        # normalize inputs from 0-255 to 0-1
        X_train/=255
        X_test/=255
        
        # one hot encode, value is 18 because i want to train more label from 10 - 17 -> totally has 18 from 0 - 17
        number_of_classes = 18

        y_train = np_utils.to_categorical(y_train, number_of_classes)
        y_test = np_utils.to_categorical(y_test, number_of_classes)
        
        # create model
        model = Sequential()
        model.add(Conv2D(32, (5, 5), input_shape=(X_train.shape[1], X_train.shape[2], 1), activation='relu'))
        model.add(MaxPooling2D(pool_size=(2, 2)))
        model.add(Conv2D(32, (3, 3), activation='relu'))
        model.add(MaxPooling2D(pool_size=(2, 2)))
        model.add(Dropout(0.5))
        model.add(Flatten())
        model.add(Dense(128, activation='relu'))
        model.add(Dropout(0.5))
        model.add(Dense(number_of_classes, activation='softmax'))
        
        # Compile model
        model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])
        # Fit the model
        model.fit(X_train, y_train, validation_data=(X_test, y_test), epochs=5)
        
        if writeFile is True:
            model.save('traineddata/traineddata.model')
    else:
        # Load the model
        model = tf.keras.models.load_model('traineddata/traineddata.model')
    
    img = cv.imread('trainmore/12.png',0)
    # img = cv.GaussianBlur(img, (1,1), 0)
    img = cv.resize(img, (28,28))
    img = np.array(img)
    img = np.invert(img)
    plt.imshow(img)
    plt.show()
    
    img = img.reshape(1,28,28,1)
    prediction = model.predict(img)
    result = np.argmax(prediction)
    print(prediction)
    print(result)
    
    return model
        
def load_images_to_data(image_label, image_directory, features_data, label_data):
    list_of_files = os.listdir(image_directory)
    for file in list_of_files:
        arr_file = file.split('.')
        image_label = arr_file[0]
        image_file_name = os.path.join(image_directory, file)
        if ".png" in image_file_name:
            img = cv.imread(image_file_name, 0)
            img = cv.resize(img, (28,28))
            im2arr = np.array(img)
            im2arr = np.invert(im2arr)
            # plt.imshow(im2arr)
            # plt.show()
            im2arr = im2arr.reshape(1,28,28,1)
            features_data = np.append(features_data, im2arr, axis=0)
            label_data = np.append(label_data, [image_label], axis=0)
    return features_data, label_data

Answer 1

我个人要做的是训练一个带有数字识别器的数字检测器，这与你知道的 YOLO 的想法相同。首先，您训练检测器以检测图像中的所有数字，然后添加在 MNIST 上训练的模型以对每个检测到的数字进行分类。