Keras 中的 Segnet：新數組的總大小必須保持不變錯誤

Question

我正在用 Python 實現 Segnet。 以下是代碼。

img_w = 480
img_h = 360
pool_size = 2

def build_model(img_w, img_h, pool_size):
    n_labels = 12

    kernel = 3

    encoding_layers = [
        Conv2D(64, (kernel, kernel), input_shape=(img_h, img_w, 3), padding='same'),
        BatchNormalization(),
        Activation('relu'),
        Convolution2D(64, (kernel, kernel), padding='same'),
        BatchNormalization(),
        Activation('relu'),
        MaxPooling2D(pool_size = (pool_size,pool_size)),

        Convolution2D(128, (kernel, kernel), padding='same'),
        BatchNormalization(),
        Activation('relu'),
        Convolution2D(128, (kernel, kernel), padding='same'),
        BatchNormalization(),
        Activation('relu'),
        MaxPooling2D(pool_size = (pool_size,pool_size)),

        Convolution2D(256, (kernel, kernel), padding='same'),
        BatchNormalization(),
        Activation('relu'),
        Convolution2D(256, (kernel, kernel), padding='same'),
        BatchNormalization(),
        Activation('relu'),
        Convolution2D(256, (kernel, kernel), padding='same'),
        BatchNormalization(),
        Activation('relu'),
        MaxPooling2D(pool_size = (pool_size,pool_size)),

        Convolution2D(512, (kernel, kernel), padding='same'),
        BatchNormalization(),
        Activation('relu'),
        Convolution2D(512, (kernel, kernel), padding='same'),
        BatchNormalization(),
        Activation('relu'),
        Convolution2D(512, (kernel, kernel), padding='same'),
        BatchNormalization(),
        Activation('relu'),
        MaxPooling2D(pool_size = (pool_size,pool_size)),

        Convolution2D(512, (kernel, kernel), padding='same'),
        BatchNormalization(),
        Activation('relu'),
        Convolution2D(512, (kernel, kernel), padding='same'),
        BatchNormalization(),
        Activation('relu'),
        Convolution2D(512, (kernel, kernel), padding='same'),
        BatchNormalization(),
        Activation('relu'),
        MaxPooling2D(pool_size = (pool_size,pool_size)),
    ]

    autoencoder = models.Sequential()
    autoencoder.encoding_layers = encoding_layers

    for l in autoencoder.encoding_layers:
        autoencoder.add(l)

    decoding_layers = [
        UpSampling2D(),
        Convolution2D(512, (kernel, kernel), padding='same'),
        BatchNormalization(),
        Activation('relu'),
        Convolution2D(512, (kernel, kernel), padding='same'),
        BatchNormalization(),
        Activation('relu'),
        Convolution2D(512, (kernel, kernel), padding='same'),
        BatchNormalization(),
        Activation('relu'),

        UpSampling2D(),
        Convolution2D(512, (kernel, kernel), padding='same'),
        BatchNormalization(),
        Activation('relu'),
        Convolution2D(512, (kernel, kernel), padding='same'),
        BatchNormalization(),
        Activation('relu'),
        Convolution2D(256, (kernel, kernel), padding='same'),
        BatchNormalization(),
        Activation('relu'),

        UpSampling2D(),
        Convolution2D(256, (kernel, kernel), padding='same'),
        BatchNormalization(),
        Activation('relu'),
        Convolution2D(256, (kernel, kernel), padding='same'),
        BatchNormalization(),
        Activation('relu'),
        Convolution2D(128, (kernel, kernel), padding='same'),
        BatchNormalization(),
        Activation('relu'),

        UpSampling2D(),
        Convolution2D(128, (kernel, kernel), padding='same'),
        BatchNormalization(),
        Activation('relu'),
        Convolution2D(64, (kernel, kernel), padding='same'),
        BatchNormalization(),
        Activation('relu'),

        UpSampling2D(),
        Convolution2D(64, (kernel, kernel), padding='same'),
        BatchNormalization(),
        Activation('relu'),
        Convolution2D(n_labels, (1, 1), padding='valid', activation="sigmoid"),
        BatchNormalization(),
    ]
    autoencoder.decoding_layers = decoding_layers
    for l in autoencoder.decoding_layers:
        autoencoder.add(l)

    autoencoder.add(Reshape((n_labels, img_h * img_w)))
    autoencoder.add(Permute((2, 1)))
    autoencoder.add(Activation('softmax'))



    return autoencoder

model = build_model(img_w, img_h, pool_size)

但它返回給我錯誤。

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
<ipython-input-21-051f06a53a14> in <module>()
----> 1 model = build_model(img_w, img_h, pool_size)

<ipython-input-20-c37fd94c8641> in build_model(img_w, img_h, pool_size)
    119         autoencoder.add(l)
    120 
--> 121     autoencoder.add(Reshape((n_labels, img_h * img_w)))
    122     autoencoder.add(Permute((2, 1)))
    123     autoencoder.add(Activation('softmax'))
    ValueError: total size of new array must be unchanged

我看不出錯誤的任何原因。 當我將 img_w 和 img_h 更改為 256 時，此錯誤已解決，但問題是這不是圖像大小或原始數據集，因此我無法使用它。 如何解決這個問題？

Answer 1

問題是您正在執行(2, 2)下采樣 5 次，所以讓我們跟蹤形狀：

(360, 480) -> (180, 240) -> (90, 120) -> (45, 60) -> (22, 30) -> (11, 15)

現在上采樣：

(11, 15) -> (22, 30) -> (44, 60) -> (88, 120) -> (176, 240) -> (352, 480)

因此，當您嘗試使用原始形狀reshape輸出時 - 由於模型不匹配而引發了問題。

可能的解決方案：

1. 調整圖像的大小，使兩個輸入尺寸都可以被32整除（例如(352, 480)或(384, 480) 。

2. 在第三次上采樣后添加ZeroPadding2D(((1, 0), (0, 0)))以將形狀從(44, 60)更改為(45, 60) ，這將使您的網絡以良好的輸出形狀結束。

其他問題：

請找出最后MaxPooling2D之后第一Upsampling2D 。 這可能是一個問題，因為這是您的網絡無用的瓶頸。