models.fusion package

Submodules

models.fusion.lightning_module module

LightningModule wrappers for feature fusion U-Net with attention mechanism and URR.

class models.fusion.lightning_module.FourStreamAttentionLightningModule(batch_size: int, metric: Metric | None = None, loss: Module | str | None = None, model_type: ModelType = ModelType.UNET, encoder_name: str = 'resnet34', encoder_depth: int = 5, encoder_weights: str | None = 'imagenet', in_channels: int = 3, classes: int = 1, num_frames: Literal[5, 10, 15, 20, 30] = 5, weights_from_ckpt_path: str | None = None, optimizer: type[Optimizer] | str = 'adamw', optimizer_kwargs: dict[str, Any] | None = None, scheduler: type[LRScheduler] | str = 'gradual_warmup_scheduler', scheduler_kwargs: dict[str, Any] | None = None, multiplier: int = 2, total_epochs: int = 50, learning_rate: float = 0.0003, dl_classification_mode: ClassificationMode = ClassificationMode.MULTICLASS_MODE, eval_classification_mode: ClassificationMode = ClassificationMode.MULTICLASS_MODE, residual_mode: ResidualMode = ResidualMode.SUBTRACT_NEXT_FRAME, loading_mode: LoadingMode = LoadingMode.RGB, dump_memory_snapshot: bool = False, flat_conv: bool = False, unet_activation: str | None = None, attention_reduction: Literal['sum', 'prod', 'cat', 'weighted', 'weighted_learnable'] = 'sum', attention_only: bool = False, dummy_predict: DummyPredictMode = DummyPredictMode.NONE, temporal_conv_type: TemporalConvolutionalType = TemporalConvolutionalType.ORIGINAL, metric_mode: MetricMode = MetricMode.INCLUDE_EMPTY_CLASS, metric_div_zero: float = 1.0, single_attention_instance: bool = False, **kwargs: Mapping)

Bases: CommonModelMixin

LightningModule wrapper for feature fusion guided U-Net with URR.

batch_size: Batch size of dataloader.

in_channels: Number of image channels.

classes: int: Number of segmentation classes.

num_frames: Number of frames used.

dump_memory_snapshot: Whether to dump a memory snapshot.

dummy_predict: Whether to simply return the ground truth for visualisation.

residual_mode: Residual frames generation mode.

optimizer: type[Optimizer] | Optimizer | str: Optimizer for training.

optimizer_kwargs: dict[str, Any]: Optimizer kwargs.

scheduler: type[LRScheduler] | LRScheduler | str: Scheduler for training.

scheduler_kwargs: dict[str, Any]: Scheduler kwargs.

loading_mode: Image loading mode.

multiplier: Learning rate multiplier.

total_epochs: int: Number of total epochs for training.

learning_rate: Learning rate for training.

flat_conv: Whether to use flat temporal convolutions.

single_attention_instance: Whether to only use 1 attention module to compute cross-attention embeddings.

weights_from_ckpt_path: Model checkpoint path to load weights from.

forward(xs: Tensor, xr: Tensor, xt: Tensor, xta_mask: Tensor, xl: Tensor) → tuple[Tensor, Tensor, Tensor]

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

log_metrics(prefix) → None

Implement shared metric logging epoch end here.

Note: This is to prevent circular imports with the logging module.

training_step(batch: tuple[Tensor, Tensor, Tensor, Tensor, Tensor, Tensor, str], batch_idx: int)

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.

validation_step(batch: tuple[Tensor, Tensor, Tensor, Tensor, Tensor, Tensor, str], batch_idx: int)

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.
    ...

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.

test_step(batch: tuple[Tensor, Tensor, Tensor, Tensor, Tensor, Tensor, str], batch_idx: int)

Operates on a single batch of data from the test set. In this step you’d normally generate examples or calculate anything of interest such as 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 test dataloader:
def test_step(self, batch, batch_idx): ...


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

Examples:

# CASE 1: A single test dataset
def test_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)
    test_acc = torch.sum(y == labels_hat).item() / (len(y) * 1.0)

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

If you pass in multiple test dataloaders, test_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 test dataloaders
def test_step(self, batch, batch_idx, dataloader_idx=0):
    # dataloader_idx tells you which dataset this is.
    ...

Note

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

Note

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

predict_step(batch: tuple[Tensor, Tensor, Tensor, Tensor, Tensor, Tensor, str | list[str]], batch_idx: int, dataloader_idx: int = 0)

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)

configure_optimizers()

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 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.

# The ReduceLROnPlateau scheduler requires a monitor
def configure_optimizers(self):
    optimizer = Adam(...)
    return {
        "optimizer": optimizer,
        "lr_scheduler": {
            "scheduler": ReduceLROnPlateau(optimizer, ...),
            "monitor": "metric_to_track",
            "frequency": "indicates how often the metric is updated",
            # If "monitor" references validation metrics, then "frequency" should be set to a
            # multiple of "trainer.check_val_every_n_epoch".
        },
    }


# In the case of two optimizers, only one using the ReduceLROnPlateau scheduler
def configure_optimizers(self):
    optimizer1 = Adam(...)
    optimizer2 = SGD(...)
    scheduler1 = ReduceLROnPlateau(optimizer1, ...)
    scheduler2 = LambdaLR(optimizer2, ...)
    return (
        {
            "optimizer": optimizer1,
            "lr_scheduler": {
                "scheduler": scheduler1,
                "monitor": "metric_to_track",
            },
        },
        {"optimizer": optimizer2, "lr_scheduler": scheduler2},
    )

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.

class models.fusion.lightning_module.ThreeStreamAttentionLightningModule(batch_size: int, metric: Metric | None = None, loss: Module | str | None = None, model_type: ModelType = ModelType.UNET, encoder_name: str = 'resnet34', encoder_depth: int = 5, encoder_weights: str | None = 'imagenet', in_channels: int = 3, classes: int = 1, num_frames: Literal[5, 10, 15, 20, 30] = 5, weights_from_ckpt_path: str | None = None, optimizer: type[Optimizer] | str = 'adamw', optimizer_kwargs: dict[str, Any] | None = None, scheduler: type[LRScheduler] | str = 'gradual_warmup_scheduler', scheduler_kwargs: dict[str, Any] | None = None, multiplier: int = 2, total_epochs: int = 50, learning_rate: float = 0.0003, dl_classification_mode: ClassificationMode = ClassificationMode.MULTICLASS_MODE, eval_classification_mode: ClassificationMode = ClassificationMode.MULTICLASS_MODE, residual_mode: ResidualMode = ResidualMode.SUBTRACT_NEXT_FRAME, loading_mode: LoadingMode = LoadingMode.RGB, dump_memory_snapshot: bool = False, flat_conv: bool = False, unet_activation: str | None = None, attention_reduction: Literal['sum', 'prod', 'cat', 'weighted', 'weighted_learnable'] = 'sum', attention_only: bool = False, dummy_predict: DummyPredictMode = DummyPredictMode.NONE, temporal_conv_type: TemporalConvolutionalType = TemporalConvolutionalType.ORIGINAL, metric_mode: MetricMode = MetricMode.INCLUDE_EMPTY_CLASS, metric_div_zero: float = 1.0, single_attention_instance: bool = False, use_stn: bool = False, **kwargs: Mapping)

Bases: CommonModelMixin

LightningModule wrapper for feature fusion guided U-Net with URR.

batch_size: Batch size of dataloader.

in_channels: Number of image channels.

classes: int: Number of segmentation classes.

num_frames: Number of frames used.

dump_memory_snapshot: Whether to dump a memory snapshot.

dummy_predict: Whether to simply return the ground truth for visualisation.

residual_mode: Residual frames generation mode.

optimizer: type[Optimizer] | Optimizer | str: Optimizer for training.

optimizer_kwargs: dict[str, Any]: Optimizer kwargs.

scheduler: type[LRScheduler] | LRScheduler | str: Scheduler for training.

scheduler_kwargs: dict[str, Any]: Scheduler kwargs.

loading_mode: Image loading mode.

multiplier: Learning rate multiplier.

total_epochs: int: Number of total epochs for training.

learning_rate: Learning rate for training.

flat_conv: Whether to use flat temporal convolutions.

single_attention_instance: Whether to only use 1 attention module to compute cross-attention embeddings.

use_stn: Whether to use a spatial transformer network to transform input images.

weights_from_ckpt_path: Model checkpoint path to load weights from.

forward(xs: Tensor, xr: Tensor, xt: Tensor, xta_mask: Tensor) → tuple[Tensor, Tensor, Tensor]

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

log_metrics(prefix) → None

Implement shared metric logging epoch end here.

Note: This is to prevent circular imports with the logging module.

training_step(batch: tuple[Tensor, Tensor, Tensor, Tensor, Tensor, str], batch_idx: int)

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.

validation_step(batch: tuple[Tensor, Tensor, Tensor, Tensor, Tensor, str], batch_idx: int)

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.
    ...

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.

test_step(batch: tuple[Tensor, Tensor, Tensor, Tensor, Tensor, str], batch_idx: int)

Operates on a single batch of data from the test set. In this step you’d normally generate examples or calculate anything of interest such as 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 test dataloader:
def test_step(self, batch, batch_idx): ...


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

Examples:

# CASE 1: A single test dataset
def test_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)
    test_acc = torch.sum(y == labels_hat).item() / (len(y) * 1.0)

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

If you pass in multiple test dataloaders, test_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 test dataloaders
def test_step(self, batch, batch_idx, dataloader_idx=0):
    # dataloader_idx tells you which dataset this is.
    ...

Note

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

Note

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

predict_step(batch: tuple[Tensor, Tensor, Tensor, Tensor, Tensor, Tensor, str | list[str]], batch_idx: int, dataloader_idx: int = 0)

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)

configure_optimizers()

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 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.

# The ReduceLROnPlateau scheduler requires a monitor
def configure_optimizers(self):
    optimizer = Adam(...)
    return {
        "optimizer": optimizer,
        "lr_scheduler": {
            "scheduler": ReduceLROnPlateau(optimizer, ...),
            "monitor": "metric_to_track",
            "frequency": "indicates how often the metric is updated",
            # If "monitor" references validation metrics, then "frequency" should be set to a
            # multiple of "trainer.check_val_every_n_epoch".
        },
    }


# In the case of two optimizers, only one using the ReduceLROnPlateau scheduler
def configure_optimizers(self):
    optimizer1 = Adam(...)
    optimizer2 = SGD(...)
    scheduler1 = ReduceLROnPlateau(optimizer1, ...)
    scheduler2 = LambdaLR(optimizer2, ...)
    return (
        {
            "optimizer": optimizer1,
            "lr_scheduler": {
                "scheduler": scheduler1,
                "monitor": "metric_to_track",
            },
        },
        {"optimizer": optimizer2, "lr_scheduler": scheduler2},
    )

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.

models.fusion.model module

Feature fusion modules.

class models.fusion.model.BERTModule(bert_type: str = 'microsoft/BiomedVLP-CXR-BERT-specialized', project_dim: int = 768)

Bases: Module

BERT submodule.

forward(input_ids: LongTensor, attention_mask: FloatTensor) → dict[str, Tensor]

Define 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.

class models.fusion.model.ThreeStreamVisionModule(encoder_name: str = 'resnet50', encoder_depth: int = 5, encoder_weights: str | None = 'imagenet', num_frames: int = 5, in_channels: int = 1, residual_mode: ResidualMode = ResidualMode.SUBTRACT_NEXT_FRAME, reduce: Literal['sum', 'prod', 'cat', 'weighted', 'weighted_learnable'] = 'sum')

Bases: Module

Three stream task vision module.

check_input_shape(x: Tensor)

Check input shape and raise an error if the shape is wrong.

Parameters:: x – Input tensor.

forward(xs: Tensor, xr: Tensor) → tuple[Sequence[Tensor], Sequence[Tensor]]

Define 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.

class models.fusion.model.FourStreamVisionModule(encoder_name: str = 'resnet50', encoder_depth: int = 5, encoder_weights: str | None = 'imagenet', num_frames: int = 5, in_channels: int = 1, residual_mode: ResidualMode = ResidualMode.SUBTRACT_NEXT_FRAME, reduce: Literal['sum', 'prod', 'cat', 'weighted', 'weighted_learnable'] = 'sum')

Bases: Module

Four stream task vision module.

check_input_shape(x: Tensor)

Check input shape and raise an error if the shape is wrong.

Parameters:: x – Input tensor.

forward(xs: Tensor, xr: Tensor, xl: Tensor) → tuple[Sequence[Tensor], Sequence[Tensor], Sequence[Tensor]]

Define 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.

class models.fusion.model.PositionalEncoding(d_model: int, dropout: float = 0.0, max_len: int = 12544, **kwargs)

Bases: Module

Positional encoding module.

forward(x: Tensor) → Tensor

Define 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.

class models.fusion.model.FusionLayer(spatial_out_channels: int, in_channels: int, output_text_len: int, input_text_len: int = 24, embed_dim: int = 768, **kwargs: Mapping)

Bases: Module

Feature fusion layer.

forward(x: Tensor, txt: Tensor) → Tensor

Define 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.

models.fusion.segmentation_model module

Feature fusion model interop with SegmentationModelsPytorch.

class models.fusion.segmentation_model.FourStreamAttentionUnet(*args, **kwargs)

Bases: SegmentationModel

U-Net with cine spatial and temporal, lge spatial, and textual feature fusion.

initialize() → None

property encoder: Get the encoder of the model.

forward(xs: Tensor, xr: Tensor, xt: Tensor, xt_a_mask: Tensor, xl: Tensor) → Tensor: Sequentially pass x trough model`s encoder, decoder and heads

class models.fusion.segmentation_model.ThreeStreamAttentionUnet(*args, **kwargs)

Bases: SegmentationModel

U-Net with LGE spatial, cine residual, and textual feature fusion.

initialize() → None

forward(xs: Tensor, xr: Tensor, xt: Tensor, xt_a_mask: Tensor) → Tensor: Sequentially pass x trough model`s encoder, decoder and heads

Module contents

Fusion models for the initial concept for enhancement.

models.fusion package

Subpackages

Submodules

models.fusion.lightning_module module

models.fusion.model module

models.fusion.segmentation_model module

Module contents