.rst .pdf

mlreco.models.singlep module

class mlreco.models.singlep.ParticleImageClassifier(cfg, name='particle_image_classifier')

Bases: torch.nn.modules.module.Module

MODULES = ['particle_image_classifier', 'network_base', 'mink_encoder']

__init__(cfg, name='particle_image_classifier')

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

forward(input)

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.singlep'

training: bool

class mlreco.models.singlep.DUQParticleClassifier(cfg, name='duq_particle_classifier')

Bases: mlreco.models.singlep.ParticleImageClassifier

Uncertainty Estimation Using a Single Deep Deterministic Neural Network https://arxiv.org/pdf/2003.02037.pdf Joost van Amersfoort, Lewis Smith, Yee Whye Teh, Yarin Gal.

Pytorch Implementation for SparseConvNets with MinkowskiEngine backend.

MODULES = ['network_base', 'particle_image_classifier', 'mink_encoder']

__init__(cfg, name='duq_particle_classifier')

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

embed(x)

bilinear(z)

forward(input)

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.

update_buffers()

__module__ = 'mlreco.models.singlep'

training: bool

class mlreco.models.singlep.EvidentialParticleClassifier(cfg, name='evidential_image_classifier')

Bases: mlreco.models.singlep.ParticleImageClassifier

MODULES = ['network_base', 'particle_image_classifier', 'mink_encoder']

__init__(cfg, name='evidential_image_classifier')

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

forward(input)

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.singlep'

training: bool

class mlreco.models.singlep.BayesianParticleClassifier(cfg, name='bayesian_particle_classifier')

Bases: torch.nn.modules.module.Module

MODULES = ['network_base', 'mcdropout_encoder']

__init__(cfg, name='bayesian_particle_classifier')

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

evidential_forward(input)

mc_forward(input, num_samples=None)

standard_forward(input, verbose=False)

forward(input)

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.singlep'

training: bool

class mlreco.models.singlep.ParticleTypeLoss(cfg, name='particle_type_loss')

Bases: torch.nn.modules.module.Module

__init__(cfg, name='particle_type_loss')

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

forward(out, type_labels)

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.singlep'

training: bool

class mlreco.models.singlep.MultiLabelCrossEntropy(cfg, name='duq_particle_classifier')

Bases: torch.nn.modules.module.Module

__init__(cfg, name='duq_particle_classifier')

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

static calc_gradient_penalty(x, y_pred)

Code From the DUQ main Github Repository: https://github.com/y0ast/deterministic-uncertainty-quantification

Author: Joost van Amersfoort

forward(out, type_labels)

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.singlep'

training: bool

class mlreco.models.singlep.EvidentialLearningLoss(cfg, name='evidential_learning_loss')

Bases: torch.nn.modules.module.Module

__init__(cfg, name='evidential_learning_loss')

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

__module__ = 'mlreco.models.singlep'

training: bool

forward(out, type_labels, iteration=0)

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.

mlreco.models.factories module mlreco.models.spice module