bob.ip.binseg.engine.ssltrainer

Functions

guess_labels(unlabelled_images, model)	Calculate the average predictions by 2 augmentations: horizontal and vertical flips
linear_rampup(current[, rampup_length])	slowly ramp-up lambda_u
mix_up(alpha, input, target, …)	Applies mix up as described in [MIXMATCH_19].
run(model, data_loader, valid_loader, …)	Fits an FCN model using semi-supervised learning and saves it to disk.
sharpen(x, T)
square_rampup(current[, rampup_length])	slowly ramp-up lambda_u

bob.ip.binseg.engine.ssltrainer.sharpen(x, T)

bob.ip.binseg.engine.ssltrainer.mix_up(alpha, input, target, unlabelled_input, unlabled_target)

Applies mix up as described in [MIXMATCH_19].

Parameters

• alpha (float) –

• input (torch.Tensor) –

• target (torch.Tensor) –

• unlabelled_input (torch.Tensor) –

• unlabled_target (torch.Tensor) –

Returns

Return type

list

bob.ip.binseg.engine.ssltrainer.square_rampup(current, rampup_length=16)

slowly ramp-up lambda_u

Parameters

• current (int) – current epoch

• rampup_length (int, optional) – how long to ramp up, by default 16

Returns

factor – ramp up factor

Return type

float

bob.ip.binseg.engine.ssltrainer.linear_rampup(current, rampup_length=16)

slowly ramp-up lambda_u

Parameters

• current (int) – current epoch

• rampup_length (int, optional) – how long to ramp up, by default 16

Returns

factor – ramp up factor

Return type

float

bob.ip.binseg.engine.ssltrainer.guess_labels(unlabelled_images, model)

Calculate the average predictions by 2 augmentations: horizontal and vertical flips

Parameters

• unlabelled_images (torch.Tensor) – [n,c,h,w]

• target (torch.Tensor) –

Returns

shape – [n,c,h,w]

Return type

torch.Tensor

bob.ip.binseg.engine.ssltrainer.run(model, data_loader, valid_loader, optimizer, criterion, scheduler, checkpointer, checkpoint_period, device, arguments, output_folder, rampup_length)

Fits an FCN model using semi-supervised learning and saves it to disk.

This method supports periodic checkpointing and the output of a CSV-formatted log with the evolution of some figures during training.

Parameters

• model (torch.nn.Module) – Network (e.g. driu, hed, unet)

• data_loader (torch.utils.data.DataLoader) – To be used to train the model

• valid_loader (torch.utils.data.DataLoader) – To be used to validate the model and enable automatic checkpointing. If set to None, then do not validate it.

• optimizer (torch.optim) –

• criterion (torch.nn.modules.loss._Loss) – loss function

• scheduler (torch.optim) – learning rate scheduler

• checkpointer (bob.ip.binseg.utils.checkpointer.Checkpointer) – checkpointer implementation

• checkpoint_period (int) – save a checkpoint every n epochs. If set to 0 (zero), then do not save intermediary checkpoints

• device (str) – device to use 'cpu' or cuda:0

• arguments (dict) – start and end epochs

• output_folder (str) – output path

• rampup_length (int) – rampup epochs