Skip to content

RegNet#

eqxvision.models.RegNet #

A simple port of torchvision.models.regnet

__init__(self, block_params: BlockParams, num_classes: int = 1000, stem_width: int = 32, stem_type: Optional[equinox.module.Module] = None, block_type: Optional[equinox.module.Module] = None, norm_layer: Optional[equinox.module.Module] = None, activation: Optional[Callable] = None, *, key: jax.random.PRNGKey = None) #

Arguments:

  • block_params: Configuration for the building blocks of the network
  • num_classes: Number of classes in the classification task. Also controls the final output shape (num_classes,). Defaults to 1000
  • stem_width: Width of stems in the model
  • stem_type: Block type for the stems
  • block_type: Type of block to be used in building the model
  • norm_layer: Normalisation to be applied on the inputs. Defaults to BatchNorm
  • activation: Activation to be applied to the intermediate outputs. Defaults to jax.nn.relu
  • key: - key: A jax.random.PRNGKey used to provide randomness for parameter initialisation. (Keyword only argument.)
__call__(self, x: Array, *, key: jax.random.PRNGKey) -> Array #

Arguments:

  • x: The input. Should be a JAX array with 3 channels
  • key: Required parameter. Utilised by few layers such as Dropout or DropPath

eqxvision.models.regnet_x_400mf(torch_weights: str = None, **kwargs: Any) -> RegNet #

Constructs a RegNetX_400MF architecture from Designing Network Design Spaces.

Arguments:

  • torch_weights: A Path or URL for the PyTorch weights. Defaults to None

eqxvision.models.regnet_x_800mf(torch_weights: str = None, **kwargs: Any) -> RegNet #

Constructs a RegNetX_800MF architecture from Designing Network Design Spaces.

Arguments:

  • torch_weights: A Path or URL for the PyTorch weights. Defaults to None

eqxvision.models.regnet_x_8gf(torch_weights: str = None, **kwargs: Any) -> RegNet #

Constructs a RegNetX_8GF architecture from Designing Network Design Spaces.

Arguments:

  • torch_weights: A Path or URL for the PyTorch weights. Defaults to None

eqxvision.models.regnet_x_16gf(torch_weights: str = None, **kwargs: Any) -> RegNet #

Constructs a RegNetX_16GF architecture from Designing Network Design Spaces.

Arguments:

  • torch_weights: A Path or URL for the PyTorch weights. Defaults to None

eqxvision.models.regnet_x_32gf(torch_weights: bool = False, **kwargs: Any) -> RegNet #

Constructs a RegNetX_32GF architecture from Designing Network Design Spaces.

Arguments:

  • torch_weights: A Path or URL for the PyTorch weights. Defaults to None

eqxvision.models.regnet_x_1_6gf(torch_weights: str = None, **kwargs: Any) -> RegNet #

Constructs a RegNetX_1.6GF architecture from Designing Network Design Spaces.

Arguments:

  • torch_weights: A Path or URL for the PyTorch weights. Defaults to None

eqxvision.models.regnet_x_3_2gf(torch_weights: str = None, **kwargs: Any) -> RegNet #

Constructs a RegNetX_3.2GF architecture from Designing Network Design Spaces.

Arguments:

  • torch_weights: A Path or URL for the PyTorch weights. Defaults to None

eqxvision.models.regnet_y_400mf(torch_weights: str = None, **kwargs: Any) -> RegNet #

Constructs a RegNetY_400MF architecture from Designing Network Design Spaces.

Arguments:

  • torch_weights: A Path or URL for the PyTorch weights. Defaults to None

eqxvision.models.regnet_y_800mf(torch_weights: str = None, **kwargs: Any) -> RegNet #

Constructs a RegNetY_800MF architecture from Designing Network Design Spaces.

Arguments:

  • torch_weights: A Path or URL for the PyTorch weights. Defaults to None

eqxvision.models.regnet_y_8gf(torch_weights: str = None, **kwargs: Any) -> RegNet #

Constructs a RegNetY_8GF architecture from Designing Network Design Spaces.

Arguments:

  • torch_weights: A Path or URL for the PyTorch weights. Defaults to None

eqxvision.models.regnet_y_16gf(torch_weights: str = None, **kwargs: Any) -> RegNet #

Constructs a RegNetY_16GF architecture from Designing Network Design Spaces.

Arguments:

  • torch_weights: A Path or URL for the PyTorch weights. Defaults to None

eqxvision.models.regnet_y_32gf(torch_weights: str = None, **kwargs: Any) -> RegNet #

Constructs a RegNetY_32GF architecture from Designing Network Design Spaces.

Arguments:

  • torch_weights: A Path or URL for the PyTorch weights. Defaults to None

eqxvision.models.regnet_y_1_6gf(torch_weights: str = None, **kwargs: Any) -> RegNet #

Constructs a RegNetY_1.6GF architecture from Designing Network Design Spaces.

Arguments:

  • torch_weights: A Path or URL for the PyTorch weights. Defaults to None

eqxvision.models.regnet_y_3_2gf(torch_weights: str = None, **kwargs: Any) -> RegNet #

Constructs a RegNetY_3.2GF architecture from Designing Network Design Spaces.

Arguments:

  • torch_weights: A Path or URL for the PyTorch weights. Defaults to None