Artifact Removal¶

class best.dbs.artifact_removal.dataset.RCSDataset(path='/mnt/Helium/filip/Projects/2020_Sleep_Analysis/2022_sleep_architecture_only/M1/M1_nostim_1575446160.0/M1_nostim_1575446160_250Hz_filtered_e0-e3_e12-e13_e4-e5_e8-e11.mefd')¶

class best.dbs.artifact_removal.dataset.StimArtifactDataset(sig_len=60, use_models=['MultiCenteriEEG_pathology', 'MultiCenteriEEG_physiology'], fs=500, use_artifacts=['RCS'], device='cpu')¶

property device¶

to(device)¶

class best.dbs.artifact_removal.dataset.StimArtifactDataset_RCS(sig_len=60, use_models=[], fs=500, use_artifacts=['RCS'], device='cpu')¶

property device¶

to(device)¶

class best.dbs.artifact_removal.model.dbs_artifact_removal_network(n_filters=64, fs=500)¶

property device¶

forward(x_inp)¶

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.

training = None¶

class best.dbs.artifact_removal.model.dbs_artifact_removal_network_(n_filters=64, fs=500)¶

property device¶

forward(x_inp)¶

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.

training = None¶

class best.dbs.artifact_removal.model.dbs_artifact_removal_network_v2¶

property device¶

forward(x_inp)¶

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.

training = None¶

class best.dbs.artifact_removal.trainer.Trainer(config)¶

do_epoch()¶

plot_to_file(x_orig, x_art, x_rec, y_art, yy_art)¶

print_losses_to_file(losses)¶

save_model()¶

train()¶

class best.dbs.artifact_removal.trainer.TrainerLight(config)¶

do_epoch()¶

plot_to_file(x_orig, x_art, x_rec, y_art)¶

print_losses_to_file(losses)¶

save_model()¶

train()¶

class best.dbs.artifact_removal.trainer.TrainerUpgrade(config)¶

do_epoch()¶

plot_to_file(x_orig, x_art, x_rec, y_art, yy_art)¶

print_losses_to_file(losses)¶

save_model()¶

train()¶

class best.dbs.artifact_removal.trainer.Trainer_(config)¶

do_epoch()¶

plot_to_file(x_orig, x_art, x_rec, y_art, yy_art)¶

print_losses_to_file(losses)¶

save_model()¶

train()¶

best.dbs.artifact_removal._remove_stimulation_artifacts.remove_artifacts(x, fs, cuda=3)¶

Parameters

x – numpy array with shape[N]
fs – 250 or 500 Hz
cuda – integer denoting id of Cuda to be used, alternatively ‘cpu’ is accepted as well

Returns