Kernel size
Kernel size is the extent of a convolutional, pooling, or analogous sliding window along each input dimension.
Evolutionary Deep Learning: Genetic algorithms and neural networks
Kernel size is the spatial extent of a convolutional filter applied by a convolutional layer.
- How it is used
- Uses kernel size to configure convolutional and pooling layers while exploring CNN architectures and evolutionary optimization.
- How deep it goes
- Defines kernel dimensions through Keras code, gene encodings, configured ranges, and mutation logic without developing convolution mathematics.
- Examples
- Uses 3×3 Conv2D layers, encoded dimensions such as 5 and 2, and experiments that vary kernel, pooling, or layer settings.
- What you can do after
- Apply it via Keras and evolutionary mutation code, then rerun models to inspect architecture and training-result changes.
Read how Evolutionary Deep Learning: Genetic algorithms and neural networks explains kernel size →
Learn Generative AI with PyTorch
The spatial size of a convolutional kernel or receptive window applied to an input.
- How it is used
- Uses kernel size to configure convolutional filters for extracting spatial features in image-generation and computer-vision models.
- How deep it goes
- Defines kernel size, relates it to stride and padding, and works through a fixed-weight 2 × 2 convolution without deriving output formulas.
- Examples
- Demonstrates a 2 × 2 kernel scanning a 3 × 3 grayscale tensor with stride 1, zero padding, fixed weights, and zero bias.
- What you can do after
- Build a PyTorch Conv2d layer by selecting kernel_size, stride, channels, padding, and optional bias.
Read how Learn Generative AI with PyTorch explains kernel size →
Math and Architectures of Deep Learning
The spatial extent of the sliding window used by a pooling layer, determining the patch size summarized at each step.
- How it is used
- Controls the local receptive patch and contributes to the spatial size of convolutional or pooling outputs.
- How deep it goes
- Explains kernel size through sliding one-dimensional and two-dimensional windows, but defers detailed output-size formulas.
- Examples
- Uses size-3 one-dimensional convolution and 2 × 2 max- and average-pooling examples.
- What you can do after
- Apply it via PyTorch MaxPool2d or AvgPool2d by selecting kernel_size and stride values.
Read how Math and Architectures of Deep Learning explains kernel size →
Inside Deep Learning: Math, Algorithms, Models
The spatial extent of a convolutional filter.
- How it is used
- Controls the spatial neighborhood examined by convolution, pooling, and transposed-convolution layers.
- How deep it goes
- Defines window size and relates it to stride and padding, but does not derive convolution output-size formulas.
- Examples
- Uses K=3 convolutions, K=2 and K=3 pooling windows, and GAN settings k_size=5 and k_size_t=4.
- What you can do after
- Apply it via PyTorch by selecting kernel size alongside padding and stride in convolutional and pooling layers.
Read how Inside Deep Learning: Math, Algorithms, Models explains kernel size →
Related concepts
- convolutional layeruses · A convolutional layer is a neural-network layer that applies shared learned filters to local regions of structured input to produce feature representations.
- max poolingused by · Max pooling is an aggregation operation that replaces each local or grouped region with its maximum value, often to downsample a representation.
- poolingused by · Pooling is an operation that aggregates values or representations across regions or axes to produce a reduced or summarized representation.
- hyperparameterkind of · A hyperparameter is a setting chosen outside ordinary parameter learning that controls a model, algorithm, or learning procedure.
- paddingcontrasts with · Padding is the addition of extra space, values, or data around or within a structure to satisfy size, alignment, layout, or processing requirements.
- expressive powercontrasts with · Expressive power is the capacity of a model, language, or system to represent or convey a range and complexity of functions, patterns, distinctions, or ideas.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.