cnn convnet - taoualiw/My-Knowledge-Base GitHub Wiki
Convolutional Neural Networks (CNN)
- a class of deep neural networks most commonly applied to images
- a network which has at least one convolutional layer. A typical CNN also includes other types of layers, such as pooling layers and dense layers.
- regularized versions of multilayer perceptrons(fully connected networks) less prone to overfitting
-
Convolution: The process of applying a kernel (filter) to an image
-
Kernel / filter: A matrix which is smaller than the input, used to transform the input into chunks
-
Padding: Adding pixels of some value, usually 0, around the input image
-
Zeros padding most common for edge values
-
Pooling: The process of reducing the size of an image through downsampling.There are several types of pooling layers.
-
max-pooling :
- most common
- reducing the size of input image by summarizing regions taking their max.
- the stride in the number of pixels to slide the kernel
- average pooling
- Convolutional Layer
- Pooling Layer
- ReLU Layer
- Fully Connected Layer
- Loss Layer
Classic network architectures:
Modern network architectures
- Inception(GoogLeNet)
- ResNet
- ResNeXt
- DenseNet
⚠️ **GitHub.com Fallback** ⚠️
