cellmil.models.mil.utils

Classes

AEM([weight_initial, weight_final, ...])

Attention Entropy Maximization (AEM) to encourage diversity in attention weights.

Accuracy_Logger(n_classes)

Accuracy logger

EarlyStopping([patience, stop_epoch, verbose])

Early stops the training if validation loss doesn't improve after a given patience.

LitGeneral(model, optimizer[, loss, ...])

class cellmil.models.mil.utils.Accuracy_Logger(n_classes: int)[source]

Bases: object

Accuracy logger

__init__(n_classes: int)[source]
initialize()[source]
log(Y_hat: Tensor, Y: Tensor)[source]
log_batch(Y_hat: Tensor, Y: Tensor)[source]
get_summary(c: int)[source]
class cellmil.models.mil.utils.EarlyStopping(patience: int = 20, stop_epoch: int = 50, verbose: bool = False)[source]

Bases: object

Early stops the training if validation loss doesn’t improve after a given patience.

__init__(patience: int = 20, stop_epoch: int = 50, verbose: bool = False)[source]
Parameters:

  • patience (int) – How long to wait after last time validation loss improved. Default: 20

  • stop_epoch (int) – Earliest epoch possible for stopping

  • verbose (bool) – If True, prints a message for each validation loss improvement. Default: False

save_checkpoint(val_loss: float, model: Module, ckpt_name: str)[source]

Saves model when validation loss decrease.

class cellmil.models.mil.utils.LitGeneral(model: Module, optimizer: Optimizer, loss: Module = CrossEntropyLoss(), lr_scheduler: Optional[LRScheduler] = None, n_classes: int = 2)[source]

Bases: LightningModule

__init__(model: Module, optimizer: Optimizer, loss: Module = CrossEntropyLoss(), lr_scheduler: Optional[LRScheduler] = None, n_classes: int = 2)[source]
forward(x: Tensor) → Tensor[source]

Same as torch.nn.Module.forward().

Parameters:

  • *args – Whatever you decide to pass into the forward method.

  • **kwargs – Keyword arguments are also possible.

Returns:

Your model’s output

configure_optimizers()[source]

Choose what optimizers and learning-rate schedulers to use in your optimization. Normally you’d need one. But in the case of GANs or similar you might have multiple. Optimization with multiple optimizers only works in the manual optimization mode.

Returns:

Any of these 6 options.

  • Single optimizer.

  • List or Tuple of optimizers.

  • Two lists - The first list has multiple optimizers, and the second has multiple LR schedulers (or multiple lr_scheduler_config).

  • Dictionary, with an "optimizer" key, and (optionally) a "lr_scheduler" key whose value is a single LR scheduler or lr_scheduler_config.

  • None - Fit will run without any optimizer.

The lr_scheduler_config is a dictionary which contains the scheduler and its associated configuration. The default configuration is shown below.

lr_scheduler_config = {
    # REQUIRED: The scheduler instance
    "scheduler": lr_scheduler,
    # The unit of the scheduler's step size, could also be 'step'.
    # 'epoch' updates the scheduler on epoch end whereas 'step'
    # updates it after a optimizer update.
    "interval": "epoch",
    # How many epochs/steps should pass between calls to
    # `scheduler.step()`. 1 corresponds to updating the learning
    # rate after every epoch/step.
    "frequency": 1,
    # Metric to monitor for schedulers like `ReduceLROnPlateau`
    "monitor": "val_loss",
    # If set to `True`, will enforce that the value specified 'monitor'
    # is available when the scheduler is updated, thus stopping
    # training if not found. If set to `False`, it will only produce a warning
    "strict": True,
    # If using the `LearningRateMonitor` callback to monitor the
    # learning rate progress, this keyword can be used to specify
    # a custom logged name
    "name": None,
}

When there are schedulers in which the .step() method is conditioned on a value, such as the torch.optim.lr_scheduler.ReduceLROnPlateau scheduler, Lightning requires that the lr_scheduler_config contains the keyword "monitor" set to the metric name that the scheduler should be conditioned on.

Metrics can be made available to monitor by simply logging it using self.log('metric_to_track', metric_val) in your LightningModule.

Note

Some things to know:

  • Lightning calls .backward() and .step() automatically in case of automatic optimization.

  • If a learning rate scheduler is specified in configure_optimizers() with key "interval" (default “epoch”) in the scheduler configuration, Lightning will call the scheduler’s .step() method automatically in case of automatic optimization.

  • If you use 16-bit precision (precision=16), Lightning will automatically handle the optimizer.

  • If you use torch.optim.LBFGS, Lightning handles the closure function automatically for you.

  • If you use multiple optimizers, you will have to switch to ‘manual optimization’ mode and step them yourself.

  • If you need to control how often the optimizer steps, override the optimizer_step() hook.

training_step(batch: tuple[torch.Tensor, torch.Tensor], batch_idx: int)[source]

Here you compute and return the training loss and some additional metrics for e.g. the progress bar or logger.

Parameters:

  • batch – The output of your data iterable, normally a DataLoader.

  • batch_idx – The index of this batch.

  • dataloader_idx – The index of the dataloader that produced this batch. (only if multiple dataloaders used)

Returns:

  • Tensor - The loss tensor

  • dict - A dictionary which can include any keys, but must include the key 'loss' in the case of automatic optimization.

  • None - In automatic optimization, this will skip to the next batch (but is not supported for multi-GPU, TPU, or DeepSpeed). For manual optimization, this has no special meaning, as returning the loss is not required.

In this step you’d normally do the forward pass and calculate the loss for a batch. You can also do fancier things like multiple forward passes or something model specific.

Example:

def training_step(self, batch, batch_idx):
    x, y, z = batch
    out = self.encoder(x)
    loss = self.loss(out, x)
    return loss

To use multiple optimizers, you can switch to ‘manual optimization’ and control their stepping:

def __init__(self):
    super().__init__()
    self.automatic_optimization = False


# Multiple optimizers (e.g.: GANs)
def training_step(self, batch, batch_idx):
    opt1, opt2 = self.optimizers()

    # do training_step with encoder
    ...
    opt1.step()
    # do training_step with decoder
    ...
    opt2.step()

Note

When accumulate_grad_batches > 1, the loss returned here will be automatically normalized by accumulate_grad_batches internally.

on_train_epoch_end() → None[source]

Called in the training loop at the very end of the epoch.

To access all batch outputs at the end of the epoch, you can cache step outputs as an attribute of the LightningModule and access them in this hook:

class MyLightningModule(L.LightningModule):
    def __init__(self):
        super().__init__()
        self.training_step_outputs = []

    def training_step(self):
        loss = ...
        self.training_step_outputs.append(loss)
        return loss

    def on_train_epoch_end(self):
        # do something with all training_step outputs, for example:
        epoch_mean = torch.stack(self.training_step_outputs).mean()
        self.log("training_epoch_mean", epoch_mean)
        # free up the memory
        self.training_step_outputs.clear()
validation_step(batch: tuple[torch.Tensor, torch.Tensor], batch_idx: int)[source]

Operates on a single batch of data from the validation set. In this step you’d might generate examples or calculate anything of interest like accuracy.

Parameters:

  • batch – The output of your data iterable, normally a DataLoader.

  • batch_idx – The index of this batch.

  • dataloader_idx – The index of the dataloader that produced this batch. (only if multiple dataloaders used)

Returns:

  • Tensor - The loss tensor

  • dict - A dictionary. Can include any keys, but must include the key 'loss'.

  • None - Skip to the next batch.

# if you have one val dataloader:
def validation_step(self, batch, batch_idx): ...


# if you have multiple val dataloaders:
def validation_step(self, batch, batch_idx, dataloader_idx=0): ...

Examples:

# CASE 1: A single validation dataset
def validation_step(self, batch, batch_idx):
    x, y = batch

    # implement your own
    out = self(x)
    loss = self.loss(out, y)

    # log 6 example images
    # or generated text... or whatever
    sample_imgs = x[:6]
    grid = torchvision.utils.make_grid(sample_imgs)
    self.logger.experiment.add_image('example_images', grid, 0)

    # calculate acc
    labels_hat = torch.argmax(out, dim=1)
    val_acc = torch.sum(y == labels_hat).item() / (len(y) * 1.0)

    # log the outputs!
    self.log_dict({'val_loss': loss, 'val_acc': val_acc})

If you pass in multiple val dataloaders, validation_step() will have an additional argument. We recommend setting the default value of 0 so that you can quickly switch between single and multiple dataloaders.

# CASE 2: multiple validation dataloaders
def validation_step(self, batch, batch_idx, dataloader_idx=0):
    # dataloader_idx tells you which dataset this is.
    x, y = batch

    # implement your own
    out = self(x)

    if dataloader_idx == 0:
        loss = self.loss0(out, y)
    else:
        loss = self.loss1(out, y)

    # calculate acc
    labels_hat = torch.argmax(out, dim=1)
    acc = torch.sum(y == labels_hat).item() / (len(y) * 1.0)

    # log the outputs separately for each dataloader
    self.log_dict({f"val_loss_{dataloader_idx}": loss, f"val_acc_{dataloader_idx}": acc})

Note

If you don’t need to validate you don’t need to implement this method.

Note

When the validation_step() is called, the model has been put in eval mode and PyTorch gradients have been disabled. At the end of validation, the model goes back to training mode and gradients are enabled.

on_validation_epoch_end() → None[source]

Called in the validation loop at the very end of the epoch.

_flatten_and_log_metrics(computed: dict[str, torch.Tensor], prefix: str) → None[source]

Convert metric dictionary produced by torchmetrics into a flat dict of scalar values and log it with self.log_dict.

  • Vector/tensor metrics (e.g. per-class accuracy) are expanded into keys like {prefix}/class_{i}_acc.

  • Scalar tensors are converted to floats.

  • None values are converted to NaN to satisfy loggers that expect numeric scalars.

predict_step(batch: tuple[torch.Tensor, torch.Tensor], batch_idx: int)[source]

Step function called during predict(). By default, it calls forward(). Override to add any processing logic.

The predict_step() is used to scale inference on multi-devices.

To prevent an OOM error, it is possible to use BasePredictionWriter callback to write the predictions to disk or database after each batch or on epoch end.

The BasePredictionWriter should be used while using a spawn based accelerator. This happens for Trainer(strategy="ddp_spawn") or training on 8 TPU cores with Trainer(accelerator="tpu", devices=8) as predictions won’t be returned.

Parameters:

  • batch – The output of your data iterable, normally a DataLoader.

  • batch_idx – The index of this batch.

  • dataloader_idx – The index of the dataloader that produced this batch. (only if multiple dataloaders used)

Returns:

Predicted output (optional).

Example

class MyModel(LightningModule):

    def predict_step(self, batch, batch_idx, dataloader_idx=0):
        return self(batch)

dm = ...
model = MyModel()
trainer = Trainer(accelerator="gpu", devices=2)
predictions = trainer.predict(model, dm)
class cellmil.models.mil.utils.AEM(weight_initial: float = 0.1, weight_final: float = 0.01, annealing_epochs: int = 25)[source]

Bases: object

Attention Entropy Maximization (AEM) to encourage diversity in attention weights.

This class provides methods to compute the AEM weight based on cosine annealing and to calculate the attention entropy loss.

__init__(weight_initial: float = 0.1, weight_final: float = 0.01, annealing_epochs: int = 25)[source]
Parameters:

  • use_aem (bool) – Whether to use AEM loss

  • aem_weight_initial (float) – Initial weight for AEM loss

  • aem_weight_final (float) – Final weight for AEM loss after annealing

  • aem_annealing_epochs (int) – Number of epochs over which to anneal the AEM weight

get_negative_entropy(attention_weights: Tensor) → Tensor[source]

Calculate the negative entropy of attention weights to encourage diversity.

Parameters:

attention_weights (torch.Tensor) – Attention weights tensor of shape (K, N) or (1, N) where K is number of classes/branches and N is number of instances

Returns:

Negative entropy (to maximize entropy, we minimize negative entropy)

Return type:

torch.Tensor

get_weight(current_epoch: int) → float[source]

Calculate the current AEM weight using cosine annealing.

Parameters:

current_epoch (int) – Current training epoch

Returns:

Current AEM weight

Return type:

float

get_aem(current_epoch: int, attention_weights: Tensor) → Tensor[source]

Get the term to sum to the loss function for AEM. :param current_epoch: Current training epoch :type current_epoch: int :param attention_weights: Attention weights tensor :type attention_weights: torch.Tensor

Returns:

Term to add to the loss function for AEM

Return type:

float