mlreco.models.layers.common.cluster_pool module

class mlreco.models.layers.common.cluster_pool.ClusterPool(pooltype='max', p=2)[source]

Bases: torch.nn.modules.module.Module

PyTorch Layer that pools features by cluster assignment options:

pooltype - how to pool

‘max’ (default) - take max of features ‘sum’ - sum features ‘average’ - average features ‘pnorm’ - p-norm of features

p - norm to use for ‘pnorm’ option (default 2)

forward method:
inputs:

features tensor (e.g. from scn sparse tensor) list of clusters

output:

pytorch tensor of size # clusters x # features

__init__(pooltype='max', p=2)[source]

Initializes internal Module state, shared by both nn.Module and ScriptModule.

forward(features, cs)[source]

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

__module__ = 'mlreco.models.layers.common.cluster_pool'
training: bool
mlreco.models.layers.common.blocks module mlreco.models.layers.common.cnn_encoder module