〈Convolutional Neural Networks(CNN) #2 池化層(Pooling layer)〉的留言 https://www.brilliantcode.net/1586/convolutional-neural-networks-2-pooling-layer/ Fri, 03 Apr 2020 02:18:25 +0000 hourly 1 https://wordpress.org/?v=5.2.3 留言者: Convolutional Neural Networks(CNN) #6 Pooling in Backward pass - BrilliantCode.net https://www.brilliantcode.net/1586/convolutional-neural-networks-2-pooling-layer/#comment-618 Fri, 03 Apr 2020 02:18:25 +0000 https://www.brilliantcode.net/?p=1586#comment-618 […] 如果你對於Pooling layer的運作方式不太了解，可以先來讀這一篇：CNN #2 Pooling layer。若想要了解更多關於卷積神經網路於反向傳遞時的計算過程與其推導，強烈建議閱讀發布的這幾篇教學：CNN #4 Kernels in Backward、CNN #5 Feature maps & Biases in Backward。 […]

]]> 留言者: Andy Wang https://www.brilliantcode.net/1586/convolutional-neural-networks-2-pooling-layer/#comment-474 Sun, 24 Nov 2019 09:42:26 +0000 https://www.brilliantcode.net/?p=1586#comment-474 完全不知為啥Cloudflare不給顯示這張圖，不過我已經在重新上傳一張囉，應該是沒有問題了。謝謝您的留言。:)

]]> 留言者: sunrimii https://www.brilliantcode.net/1586/convolutional-neural-networks-2-pooling-layer/#comment-473 Sun, 24 Nov 2019 06:49:37 +0000 https://www.brilliantcode.net/?p=1586#comment-473 請問圖三是不是失效了呢?

]]>