resnet

ResNet

Overview

ResNet is a type of convolutional neural network (CNN) that is commonly used for image classification tasks. ResNet is a fully convolutional network that consists of a series of convolutional layers and pooling layers. The pooling layers are used to downsample the input while the convolutional layers are used to upsample the input. The skip connections between the pooling layers and convolutional layers allow ResNet to preserve spatial/temporal information while also allowing for faster training and inference times.

For more info, refer to the original paper Deep Residual Learning for Image Recognition.

Parameters are in helia_edge.models.resnet_params.

Functions

Name	Description
build	ResNet model from ResNetParams
generate_bottleneck_block	Generate functional bottleneck block
generate_residual_block	Generate functional residual block
resnet_layer	Generate functional ResNet model

Additions

• Enable 1D and 2D variants.

Machine-readable model

• generate_bottleneck_blockfunctionGenerate functional bottleneck block.
• generate_residual_blockfunctionGenerate functional residual block
• resnet_layerfunctionGenerate functional ResNet model.
• buildfunctionBuild a ResNet model.

function

generate_bottleneck_block

Python

Generate functional bottleneck block.

helia_edge/models/resnet.py:31

generate_bottleneck_block(
    filters: int,
    kernel_size: int | tuple[int, int] = 3,
    strides: int | tuple[int, int] = 1,
    expansion: int = 4,
    activation: str = 'relu6',
) -> keras.Layer

Generate functional bottleneck block.

Parameters

Parameters of generate_bottleneck_block

Name	Type	Default	Description
filters	int	Required	Filter size
kernel_size	int | tuple[int, int]	3	Kernel size. Defaults to 3.
strides	int | tuple[int, int]	1	Stride length. Defaults to 1.
expansion	int	4	Expansion factor. Defaults to 4.

Returns

Returns of generate_bottleneck_block

Type	Description
keras.Layer	keras.Layer: TF functional layer

function

generate_residual_block

Python

Generate functional residual block

helia_edge/models/resnet.py:75

generate_residual_block(
    filters: int,
    kernel_size: int | tuple[int, int] = 3,
    strides: int | tuple[int, int] = 1,
    activation: str = 'relu6',
) -> keras.Layer

Generate functional residual block

Parameters

Parameters of generate_residual_block

Name	Type	Default	Description
filters	int	Required	Filter size
kernel_size	int | tuple[int, int]	3	Kernel size. Defaults to 3.
strides	int | tuple[int, int]	1	Stride length. Defaults to 1.

Returns

Returns of generate_residual_block

Type	Description
keras.Layer	keras.Layer: TF functional layer

function

resnet_layer

Python

Generate functional ResNet model.

helia_edge/models/resnet.py:111

resnet_layer(x: keras.KerasTensor, params: ResNetParams) -> keras.KerasTensor

Generate functional ResNet model. Args: x (keras.KerasTensor): Inputs params (ResNetParams): Model parameters.

Returns

Returns of resnet_layer

Type	Description
keras.KerasTensor	keras.KerasTensor: Output tensor

function

build

Python

Build a ResNet model.

helia_edge/models/resnet.py:169

build(
    params: ResNetParams,
    input_shape: tuple[int | None, ...],
    *,
    batch_size: int | None = None,
    name: str | None = None,
) -> keras.Model

Build a ResNet model.

Parameters

Parameters of build

Name	Type	Default	Description
params	ResNetParams	Required	Model parameters.
input_shape	tuple[int | None, ...]	Required	Input shape without the batch axis; None for a variable axis.
batch_size	int | None	None	Static batch size; None for a dynamic batch.
name	str | None	None	Model name; the family when None.

Returns

Returns of build

Type	Description
keras.Model	keras.Model: The model, named ``resnet`` unless ``name`` is given.