InceptionResnetV2 STEM块角的实现与原论文中的不匹配？

Question

我一直在尝试将来自Keras实现的Keras实现模型摘要与他们的论文中指定的模型摘要进行比较，而当涉及到filter_concat块时，它似乎并没有显示出多少相似之处。

模型summary()的前几行如下所示。(就我的情况而言，输入改为512x512，但据我所知，它不影响每层过滤器的数量，因此我们也可以使用它们来跟踪纸质代码的翻译)：

Model: "inception_resnet_v2"
__________________________________________________________________________________________________
Layer (type)                    Output Shape         Param #     Connected to
==================================================================================================
input_1 (InputLayer)            (None, 512, 512, 3)  0
__________________________________________________________________________________________________
conv2d_1 (Conv2D)               (None, 255, 255, 32) 864         input_1[0][0]
__________________________________________________________________________________________________
batch_normalization_1 (BatchNor (None, 255, 255, 32) 96          conv2d_1[0][0]
__________________________________________________________________________________________________
activation_1 (Activation)       (None, 255, 255, 32) 0           batch_normalization_1[0][0]
__________________________________________________________________________________________________
conv2d_2 (Conv2D)               (None, 253, 253, 32) 9216        activation_1[0][0]
__________________________________________________________________________________________________
batch_normalization_2 (BatchNor (None, 253, 253, 32) 96          conv2d_2[0][0]
__________________________________________________________________________________________________
activation_2 (Activation)       (None, 253, 253, 32) 0           batch_normalization_2[0][0]
__________________________________________________________________________________________________
conv2d_3 (Conv2D)               (None, 253, 253, 64) 18432       activation_2[0][0]
__________________________________________________________________________________________________
batch_normalization_3 (BatchNor (None, 253, 253, 64) 192         conv2d_3[0][0]
__________________________________________________________________________________________________
activation_3 (Activation)       (None, 253, 253, 64) 0           batch_normalization_3[0][0]
__________________________________________________________________________________________________
max_pooling2d_1 (MaxPooling2D)  (None, 126, 126, 64) 0           activation_3[0][0]
__________________________________________________________________________________________________
conv2d_4 (Conv2D)               (None, 126, 126, 80) 5120        max_pooling2d_1[0][0]
__________________________________________________________________________________________________
batch_normalization_4 (BatchNor (None, 126, 126, 80) 240         conv2d_4[0][0]
__________________________________________________________________________________________________
activation_4 (Activation)       (None, 126, 126, 80) 0           batch_normalization_4[0][0]
__________________________________________________________________________________________________
conv2d_5 (Conv2D)               (None, 124, 124, 192 138240      activation_4[0][0]
__________________________________________________________________________________________________
batch_normalization_5 (BatchNor (None, 124, 124, 192 576         conv2d_5[0][0]
__________________________________________________________________________________________________
activation_5 (Activation)       (None, 124, 124, 192 0           batch_normalization_5[0][0]
__________________________________________________________________________________________________
max_pooling2d_2 (MaxPooling2D)  (None, 61, 61, 192)  0           activation_5[0][0]
__________________________________________________________________________________________________
conv2d_9 (Conv2D)               (None, 61, 61, 64)   12288       max_pooling2d_2[0][0]
__________________________________________________________________________________________________
batch_normalization_9 (BatchNor (None, 61, 61, 64)   192         conv2d_9[0][0]
__________________________________________________________________________________________________
activation_9 (Activation)       (None, 61, 61, 64)   0           batch_normalization_9[0][0]
__________________________________________________________________________________________________
conv2d_7 (Conv2D)               (None, 61, 61, 48)   9216        max_pooling2d_2[0][0]
__________________________________________________________________________________________________
conv2d_10 (Conv2D)              (None, 61, 61, 96)   55296       activation_9[0][0]
__________________________________________________________________________________________________
batch_normalization_7 (BatchNor (None, 61, 61, 48)   144         conv2d_7[0][0]
__________________________________________________________________________________________________
batch_normalization_10 (BatchNo (None, 61, 61, 96)   288         conv2d_10[0][0]
__________________________________________________________________________________________________
activation_7 (Activation)       (None, 61, 61, 48)   0           batch_normalization_7[0][0]
__________________________________________________________________________________________________
activation_10 (Activation)      (None, 61, 61, 96)   0           batch_normalization_10[0][0]
__________________________________________________________________________________________________
average_pooling2d_1 (AveragePoo (None, 61, 61, 192)  0           max_pooling2d_2[0][0]
__________________________________________________________________________________________________
.
.
. 
many more lines

在他们的论文的图3(下面的图)中，展示了如何为InceptionV4和InceptionResnetV2形成STEM块。在图3中，STEM块中有三个过滤器级联，但在我前面向您展示的输出中，级联似乎是顺序最大池或类似的混合(第一个连接应该出现在max_pooling2d_1之后)。它增加了过滤器的数目，正如级联应该做的那样，但是没有进行连接。过滤器似乎是按顺序放置的！有人知道这个输出是怎么回事吗？它的作用和论文中描述的一样？

作为比较，我找到了一个InceptionV4 keras实现，他们似乎在concatenate_1中为STEM块中的第一个级联做了一个filter_concat。下面是summary()的第一行的输出。

Model: "inception_v4"
__________________________________________________________________________________________________
Layer (type)                    Output Shape         Param #     Connected to
==================================================================================================
input_1 (InputLayer)            (None, 512, 512, 3)  0
__________________________________________________________________________________________________
conv2d_1 (Conv2D)               (None, 255, 255, 32) 864         input_1[0][0]
__________________________________________________________________________________________________
batch_normalization_1 (BatchNor (None, 255, 255, 32) 96          conv2d_1[0][0]
__________________________________________________________________________________________________
activation_1 (Activation)       (None, 255, 255, 32) 0           batch_normalization_1[0][0]
__________________________________________________________________________________________________
conv2d_2 (Conv2D)               (None, 253, 253, 32) 9216        activation_1[0][0]
__________________________________________________________________________________________________
batch_normalization_2 (BatchNor (None, 253, 253, 32) 96          conv2d_2[0][0]
__________________________________________________________________________________________________
activation_2 (Activation)       (None, 253, 253, 32) 0           batch_normalization_2[0][0]
__________________________________________________________________________________________________
conv2d_3 (Conv2D)               (None, 253, 253, 64) 18432       activation_2[0][0]
__________________________________________________________________________________________________
batch_normalization_3 (BatchNor (None, 253, 253, 64) 192         conv2d_3[0][0]
__________________________________________________________________________________________________
activation_3 (Activation)       (None, 253, 253, 64) 0           batch_normalization_3[0][0]
__________________________________________________________________________________________________
conv2d_4 (Conv2D)               (None, 126, 126, 96) 55296       activation_3[0][0]
__________________________________________________________________________________________________
batch_normalization_4 (BatchNor (None, 126, 126, 96) 288         conv2d_4[0][0]
__________________________________________________________________________________________________
max_pooling2d_1 (MaxPooling2D)  (None, 126, 126, 64) 0           activation_3[0][0]
__________________________________________________________________________________________________
activation_4 (Activation)       (None, 126, 126, 96) 0           batch_normalization_4[0][0]
__________________________________________________________________________________________________
concatenate_1 (Concatenate)     (None, 126, 126, 160 0           max_pooling2d_1[0][0]
                                                                 activation_4[0][0]
__________________________________________________________________________________________________
conv2d_7 (Conv2D)               (None, 126, 126, 64) 10240       concatenate_1[0][0]
__________________________________________________________________________________________________
batch_normalization_7 (BatchNor (None, 126, 126, 64) 192         conv2d_7[0][0]
__________________________________________________________________________________________________
activation_7 (Activation)       (None, 126, 126, 64) 0           batch_normalization_7[0][0]
__________________________________________________________________________________________________
conv2d_8 (Conv2D)               (None, 126, 126, 64) 28672       activation_7[0][0]
__________________________________________________________________________________________________
.
.
.
and many more lines

因此，正如本文所示，这两种体系结构都应该具有相同的第一层。还是我漏掉了什么？

编辑：编辑--我发现，来自Keras的InceptionResnetV2的实现不是，而是的InceptionResnetV2块，而是的实现(图14来自他们的论文，如下所示)。在STEM块之后，它似乎很好地跟随了InceptionResnetV2的其他块。

InceptionResnetV1的性能不如InceptionResnetV2 (图25)，所以我对使用来自V1的块而不是从keras中的完整V2表示怀疑。我将尝试从我找到的InceptionV4中删除STEM，并将InceptionResnetV2的继续。

同样的问题在tf-模型github没有解释的情况下结束了。如果有人感兴趣，我就把它留在这里：https://github.com/tensorflow/models/issues/1235

编辑2:出于某种原因，GoogleAI (盗梦空间体系结构的创建者)在发布代码时在他们的博客中显示了“Inception RESNET-v2”的图像。但是，STEM块是来自InceptionV3的块，而不是InceptionV4块，正如本文所指定的那样。所以，要么是纸错了，要么是代码由于某种原因没有跟随论文。

​

​

​

​

​

​

Answer 1

实现了类似的结果。

我刚收到一封电子邮件，证实了阿莱米的错误，他是谷歌的高级研究科学家，也是关于InceptionResnetV2代码发布的博客文章的原版出版商。似乎在内部实验期间，STEM块被切换，并且释放就像这样。

引用：

丹妮·阿兹马尔 看来你是对的。不完全确定发生了什么，但代码显然是真相的来源，因为发布的检查点是为同样发布的代码提供的。当我们开发架构的时候，我们做了大量的内部实验，我想在某个时候，茎被转换了。我不确定现在是否有时间深入挖掘，但正如我所说，释放的检查点是已发布代码的检查点，因为您可以通过运行评估管道来验证自己。我同意你的观点，似乎这是在使用原始的“盗梦空间”V1干细胞。诚挚的问候, 亚历克斯·阿莱米

我将更新这篇文章中有关这一主题的变化。

UPDATE：Christian Szegedy，也是原版报纸推特的出版商

最初的实验和模型是在DistBelief中创建的，这是一个完全不同的框架，在Tensorflow之前。 信托基金的版本是一年后增加的，可能与原来的模式有出入，但确保取得类似的结果。

因此，由于它取得了类似的结果，你的实验将大致相同。