mednet.models.detect.faster_rcnn

Faster R-CNN object detection (and classification) network architecture, from [RHGS17].

Classes

FasterRCNN([optimizer_type, ...])

Faster R-CNN object detection (and classification) network architecture, from [RHGS17].

class mednet.models.detect.faster_rcnn.FasterRCNN(optimizer_type=<class 'torch.optim.sgd.SGD'>, optimizer_arguments={'lr': 0.005, 'momentum': 0.9, 'weight_decay': 0.0005}, scheduler_type=<class 'torch.optim.lr_scheduler.StepLR'>, scheduler_arguments={'gamma': 0.1, 'step_size': 3}, model_transforms=None, augmentation_transforms=None, pretrained=False, num_classes=1, variant='mobilenetv3-small')[source]

Bases: Model

Faster R-CNN object detection (and classification) network architecture, from [RHGS17].

Parameters:
  • optimizer_type (type[Optimizer]) – The type of optimizer to use for training.

  • optimizer_arguments (dict[str, Any]) – Arguments to the optimizer after params.

  • scheduler_type (type[LRScheduler] | None) – The type of scheduler to use for training.

  • scheduler_arguments (dict[str, Any]) – Arguments to the scheduler after params.

  • model_transforms (Optional[Sequence[Callable[[Tensor], Tensor]]]) – An optional sequence of torch modules containing transforms to be applied on the input before it is fed into the network.

  • augmentation_transforms (Optional[Sequence[Callable[[Tensor], Tensor]]]) – An optional sequence of torch modules containing transforms to be applied on the input before it is fed into the network.

  • pretrained (bool) – If set to True, loads pretrained model weights during initialization, else trains a new model.

  • num_classes (int) – Number of outputs (classes) for this model. Do not account for the background (we compensate internally).

  • variant (Literal['resnet50-v1', 'resnet50-v2', 'mobilenetv3-large', 'mobilenetv3-small']) – One of the torchvision supported variants.

property num_classes: int

Number of outputs (classes) for this model.

Returns:

The number of outputs supported by this model.

Return type:

int

on_save_checkpoint(checkpoint)[source]

Perform actions during checkpoint saving (called by lightning).

Called by Lightning when saving a checkpoint to give you a chance to store anything else you might want to save. Use on_load_checkpoint() to restore what additional data is saved here.

Parameters:

checkpoint (MutableMapping[str, Any]) – The checkpoint to save.

Return type:

None

on_load_checkpoint(checkpoint)[source]

Perform actions during model loading (called by lightning).

If you saved something with on_save_checkpoint() this is your chance to restore this.

Parameters:

checkpoint (MutableMapping[str, Any]) – The loaded checkpoint.

Return type:

None

forward(images, targets=None)[source]

Forward the input tensor through the network, producing a prediction.

Parameters:
  • images (list[Tensor]) – Input images, to be analyzed or trained on.

  • targets (list[dict[str, Tensor]] | None) – Targets for the current input images. Targets should be passed only if training. It should be alist of dictionaries, each corresponding to one of the input images in images. Each dictionary should have two keys: boxes, that contains the bounding boxes associated with the image, and labels, that contains the labels associated with each bounding box.

Returns:

A list with various dictionaries, each referring to one of the

training_step(batch, batch_idx)[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.

validation_step(batch, batch_idx, dataloader_idx=0)[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.
    ...

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.

predict_step(batch, batch_idx, dataloader_idx=0)[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)