mbconv_params

Typed MBConv block parameters; importable without Keras.

Machine-readable model

• MBConvParamsclassMBConv parameters

class

MBConvParams

Python

MBConv parameters

helia_edge/layers/mbconv_params.py:6

MBConvParams()

MBConv parameters

attribute

model_config

Python

helia_edge/layers/mbconv_params.py:21

model_config = ConfigDict(frozen=True, extra='forbid')

attribute

filters

Python

helia_edge/layers/mbconv_params.py:23

filters: int = Field(..., description='# filters')

Number of filters

attribute

depth

Python

helia_edge/layers/mbconv_params.py:24

depth: int = Field(default=1, description='Layer depth')

Layer depth

attribute

ex_ratio

Python

helia_edge/layers/mbconv_params.py:25

ex_ratio: float = Field(default=1, description='Expansion ratio')

Expansion ratio

attribute

kernel_size

Python

helia_edge/layers/mbconv_params.py:26

kernel_size: int | tuple[int, int] = Field(default=3, description='Kernel size')

Kernel size

attribute

strides

Python

helia_edge/layers/mbconv_params.py:27

strides: int | tuple[int, int] = Field(default=1, description='Stride size')

Stride size

attribute

se_ratio

Python

helia_edge/layers/mbconv_params.py:28

se_ratio: float = Field(default=8, description='Squeeze Excite ratio')

Squeeze Excite ratio

attribute

droprate

Python

helia_edge/layers/mbconv_params.py:29

droprate: float = Field(default=0, description='Drop rate')

Drop rate

attribute

bn_momentum

Python

helia_edge/layers/mbconv_params.py:30

bn_momentum: float = Field(default=0.9, description='Batch normalization momentum')

Batch normalization momentum

attribute

activation

Python

helia_edge/layers/mbconv_params.py:31

activation: str = Field(default='relu6', description='Activation function')

Activation function