PyTorch toolbox to work with spherical surfaces.
Note
This page is a reference documentation. It only explains the class signature, and not how to use it. Please refer to the gallery for the big picture.
-
class
surfify.models.vgg.SphericalVGG(input_channels, cfg, n_classes, input_order=5, conv_mode='DiNe', dine_size=1, repa_size=5, repa_zoom=5, hidden_dim=4096, batch_norm=False, init_weights=True, standard_ico=False, cachedir=None)[source]¶ Spherical VGG architecture.
See also
Notes
Debuging messages can be displayed by changing the log level using
setup_logging(level='debug').-
__init__(input_channels, cfg, n_classes, input_order=5, conv_mode='DiNe', dine_size=1, repa_size=5, repa_zoom=5, hidden_dim=4096, batch_norm=False, init_weights=True, standard_ico=False, cachedir=None)[source]¶ Init class.
- Parameters
input_channels : int
the number of input channels.
cfg : list
the definition of layers where ‘M’ stands for max pooling.
num_classes : int
the number of class in the classification problem.
input_order : int, default 5
the input icosahedron order.
conv_mode : str, default ‘DiNe’
use either ‘RePa’ - Rectangular Patch convolution method or ‘DiNe’ - 1 ring Direct Neighbor convolution method.
dine_size : int, default 1
the size of the spherical convolution filter, ie. the number of neighbor rings to be considered.
repa_size : int, default 5
the size of the rectangular grid in the tangent space.
repa_zoom : int, default 5
a multiplicative factor applied to the rectangular grid in the tangent space.
hidden_dim : int, default 4096
the 2-layer classification MLP number of hidden dims.
batch_norm : bool, default False
wether or not to use batch normalization after a convolution layer.
init_weights : bool, default True
initialize network weights.
standard_ico : bool, default False
optionaly use surfify tesselation.
cachedir : str, default None
set this folder to use smart caching speedup.
-
Follow us