domid.trainers package

Submodules

domid.trainers.pretraining_GMM module

class domid.trainers.pretraining_GMM.Pretraining(model, device, loader_tr, loader_val, i_h, i_w, args)

Bases: object

__init__(model, device, loader_tr, loader_val, i_h, i_w, args)

Parameters:

• model – the model to train

• device – the device to use

• loader_tr – the training data loader

• i_h – image height

• i_w – image width

pretrain_loss(tensor_x, inject_tensor)

Parameters:

tensor_x – the input image

Returns:

the loss

GMM_fit()

During pre-training we estimate pi, mu_c, and log_sigma2_c with a GMM at the end of each epoch. After pre-training these initial parameter values are used in the calculation of the ELBO loss, and are further updated with backpropagation like all other neural network weights.

domid.trainers.pretraining_KMeans module

class domid.trainers.pretraining_KMeans.Pretraining(model, device, loader_tr, loader_val, i_h, i_w, args)

Bases: object

__init__(model, device, loader_tr, loader_val, i_h, i_w, args)

Parameters:

• model – the model to train

• device – the device to use

• loader_tr – the training data loader

• i_h – image height

• i_w – image width

pretrain_loss(tensor_x, inject_tensor)

Parameters:

tensor_x – the input image

Returns:

the loss

model_fit()

domid.trainers.pretraining_sdcn module

class domid.trainers.pretraining_sdcn.PretrainingSDCN(model, device, loader_tr, loader_val, i_h, i_w, args)

Bases: object

__init__(model, device, loader_tr, loader_val, i_h, i_w, args)

Parameters:

• model – the model to train

• device – the device to use

• loader_tr – the training data loader

• i_h – image height

• i_w – image width

pretrain_loss(tensor_x)

kmeans_cluster_assignement()

model_fit()

domid.trainers.trainer_ae module

class domid.trainers.trainer_ae.TrainerAE(successor_node=None, extend=None)

Bases: AbstractTrainer

init_business(model, task, observer, device, aconf, flag_accept=True)

Parameters:

• model – model to train

• task – task to train on

• observer – observer to notify

• device – device to use

• writer – tensorboard writer

• pretrain – whether to pretrain the model with MSE loss

• aconf – configuration parameters, including learning rate and pretrain threshold

tr_epoch(epoch)

Parameters:

epoch – epoch number

Returns:

before_tr()

check the performance of randomly initialized weight

post_tr()

after training

domid.trainers.trainer_cluster module

class domid.trainers.trainer_cluster.TrainerCluster(successor_node=None, extend=None)

Bases: AbstractTrainer

init_business(model, task, observer, device, aconf, flag_accept=True)

Parameters:

• model – model to train

• task – task to train on

• observer – observer to notify

• device – device to use

• writer – tensorboard writer

• pretrain – whether to pretrain the model with MSE loss

• aconf – configuration parameters, including learning rate and pretrain threshold

tr_epoch(epoch)

Parameters:

epoch – epoch number

Returns:

before_tr()

check the performance of randomly initialized weight

post_tr()

after training

domid.trainers.trainer_sdcn module

class domid.trainers.trainer_sdcn.TrainerSDCN(successor_node=None, extend=None)

Bases: AbstractTrainer

init_business(model, task, observer, device, aconf, flag_accept=True)

model, task, observer, device, aconf

tr_epoch(epoch)

Parameters:

epoch – epoch number

Returns:

before_tr()

check the performance of randomly initialized weight

post_tr()

after training

domid.trainers.zoo_trainer module

select trainer

class domid.trainers.zoo_trainer.TrainerChainNodeGetter(str_trainer)

Bases: object

Chain of Responsibility: node is named in pattern Trainer[XXX] where the string after ‘Trainer’ is the name to be passed to args.trainer.

__init__(str_trainer)

__init__. :param args: command line arguments

Module contents