minerva.models.nets
===================
.. py:module:: minerva.models.nets
Submodules
----------
.. toctree::
:maxdepth: 1
/autoapi/minerva/models/nets/base/index
/autoapi/minerva/models/nets/classic_ml_pipeline/index
/autoapi/minerva/models/nets/conv_autoencoders_encoders/index
/autoapi/minerva/models/nets/cpc_networks/index
/autoapi/minerva/models/nets/dcnn/index
/autoapi/minerva/models/nets/mlp/index
/autoapi/minerva/models/nets/siamese_network_wrapper/index
/autoapi/minerva/models/nets/tfc/index
/autoapi/minerva/models/nets/tnc/index
Classes
-------
.. autoapisummary::
minerva.models.nets.DeepLabV3
minerva.models.nets.MLP
minerva.models.nets.SETR_PUP
minerva.models.nets.SimpleSupervisedModel
minerva.models.nets.UNet
minerva.models.nets.WiseNet
Package Contents
----------------
.. py:class:: DeepLabV3(backbone = None, pred_head = None, loss_fn = None, learning_rate = 0.001, num_classes = 6, train_metrics = None, val_metrics = None, test_metrics = None)
Bases: :py:obj:`minerva.models.nets.base.SimpleSupervisedModel`
A DeeplabV3 with a ResNet50 backbone
References
----------
Liang-Chieh Chen, George Papandreou, Florian Schroff, Hartwig Adam.
"Rethinking Atrous Convolution for Semantic Image Segmentation", 2017
Initializes a DeepLabV3 model.
Parameters
----------
backbone: Optional[nn.Module]
The backbone network. Defaults to None.
pred_head: Optional[nn.Module]
The prediction head network. Defaults to None.
loss_fn: Optional[nn.Module]
The loss function. Defaults to None.
learning_rate: float
The learning rate for the optimizer. Defaults to 0.001.
num_classes: int
The number of classes for prediction. Defaults to 6.
train_metrics: Optional[Dict[str, Metric]]
The metrics to be computed during training. Defaults to None.
val_metrics: Optional[Dict[str, Metric]]
The metrics to be computed during validation. Defaults to None.
test_metrics: Optional[Dict[str, Metric]]
The metrics to be computed during testing. Defaults to None.
.. py:method:: _loss_func(y_hat, y)
Calculate the loss between the output and the input data.
Parameters
----------
y_hat : torch.Tensor
The output data from the forward pass.
y : torch.Tensor
The input data/label.
Returns
-------
torch.Tensor
The loss value.
.. py:method:: 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.
Return:
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.
.. code-block:: python
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
:class:`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.
.. testcode::
# 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 :class:`~lightning.pytorch.core.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 :class:`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 :meth:`optimizer_step` hook.
.. py:method:: forward(x)
Perform a forward pass with the input data on the backbone model.
Parameters
----------
x : torch.Tensor
The input data.
Returns
-------
torch.Tensor
The output data from the forward pass.
.. py:class:: MLP(layer_sizes, activation_cls = nn.ReLU, *args, **kwargs)
Bases: :py:obj:`torch.nn.Sequential`
A multilayer perceptron (MLP) implemented as a subclass of nn.Sequential.
This MLP is composed of a sequence of linear layers interleaved with ReLU activation
functions, except for the final layer which remains purely linear.
Example
-------
>>> mlp = MLP(10, 20, 30, 40)
>>> print(mlp)
MLP(
(0): Linear(in_features=10, out_features=20, bias=True)
(1): ReLU()
(2): Linear(in_features=20, out_features=30, bias=True)
(3): ReLU()
(4): Linear(in_features=30, out_features=40, bias=True)
)
Initializes the MLP with specified layer sizes.
Parameters
----------
layer_sizes : Sequence[int]
A sequence of positive integers indicating the size of each layer.
At least two integers are required, representing the input and output layers.
activation_cls : type
The class of the activation function to use between layers. Default is nn.ReLU.
*args
Additional arguments passed to the activation function.
**kwargs
Additional keyword arguments passed to the activation function.
Raises
------
AssertionError
If fewer than two layer sizes are provided or if any layer size is not a positive integer.
AssertionError
If activation_cls does not inherit from torch.nn.Module.
.. py:class:: SETR_PUP(image_size = 512, patch_size = 16, num_layers = 24, num_heads = 16, hidden_dim = 1024, mlp_dim = 4096, encoder_dropout = 0.1, num_classes = 1000, norm_layer = None, decoder_channels = 256, num_convs = 4, up_scale = 2, kernel_size = 3, align_corners = False, decoder_dropout = 0.1, conv_norm = None, conv_act = None, interpolate_mode = 'bilinear', loss_fn = None, optimizer_type = None, optimizer_params = None, train_metrics = None, val_metrics = None, test_metrics = None, aux_output = True, aux_output_layers = None, aux_weights = None, load_backbone_path = None, freeze_backbone_on_load = True, learning_rate = 0.001, loss_weights = None, original_resolution = None, head_lr_factor = 1.0, test_engine = None)
Bases: :py:obj:`lightning.pytorch.LightningModule`
SET-R model with PUP head for image segmentation.
Methods
-------
forward(x: torch.Tensor) -> torch.Tensor
Forward pass of the model.
_compute_metrics(y_hat: torch.Tensor, y: torch.Tensor, step_name: str)
Compute metrics for the given step.
_loss_func(y_hat: Union[torch.Tensor, Tuple[torch.Tensor, torch.Tensor, torch.Tensor, torch.Tensor]], y: torch.Tensor) -> torch.Tensor
Calculate the loss between the output and the input data.
_single_step(batch: torch.Tensor, batch_idx: int, step_name: str)
Perform a single step of the training/validation loop.
training_step(batch: torch.Tensor, batch_idx: int)
Perform a single training step.
validation_step(batch: torch.Tensor, batch_idx: int)
Perform a single validation step.
test_step(batch: torch.Tensor, batch_idx: int)
Perform a single test step.
predict_step(batch: torch.Tensor, batch_idx: int, dataloader_idx: Optional[int] = None)
Perform a single prediction step.
load_backbone(path: str, freeze: bool = False)
Load a pre-trained backbone.
configure_optimizers()
Configure the optimizer for the model.
create_from_dict(config: Dict) -> "SETR_PUP"
Create an instance of SETR_PUP from a configuration dictionary.
Initialize the SETR model with Progressive Upsampling Head.
Parameters
----------
image_size : Union[int, Tuple[int, int]], optional
Size of the input image, by default 512.
patch_size : int, optional
Size of the patches to be extracted from the input image, by
default 16.
num_layers : int, optional
Number of transformer layers, by default 24.
num_heads : int, optional
Number of attention heads, by default 16.
hidden_dim : int, optional
Dimension of the hidden layer, by default 1024.
mlp_dim : int, optional
Dimension of the MLP layer, by default 4096.
encoder_dropout : float, optional
Dropout rate for the encoder, by default 0.1.
num_classes : int, optional
Number of output classes, by default 1000.
norm_layer : Optional[nn.Module], optional
Normalization layer, by default None.
decoder_channels : int, optional
Number of channels in the decoder, by default 256.
num_convs : int, optional
Number of convolutional layers in the decoder, by default 4.
up_scale : int, optional
Upscaling factor for the decoder, by default 2.
kernel_size : int, optional
Kernel size for the convolutional layers, by default 3.
align_corners : bool, optional
Whether to align corners when interpolating, by default False.
decoder_dropout : float, optional
Dropout rate for the decoder, by default 0.1.
conv_norm : Optional[nn.Module], optional
Normalization layer for the convolutional layers, by default None.
conv_act : Optional[nn.Module], optional
Activation function for the convolutional layers, by default None.
interpolate_mode : str, optional
Interpolation mode, by default "bilinear".
loss_fn : Optional[nn.Module], optional
Loss function, when None defaults to nn.CrossEntropyLoss, by
default None.
optimizer_type : Optional[type], optional
Type of optimizer, by default None.
optimizer_params : Optional[Dict], optional
Parameters for the optimizer, by default None.
train_metrics : Optional[Dict[str, Metric]], optional
Metrics for training, by default None.
val_metrics : Optional[Dict[str, Metric]], optional
Metrics for validation, by default None.
test_metrics : Optional[Dict[str, Metric]], optional
Metrics for testing, by default None.
aux_output : bool, optional
Whether to use auxiliary outputs, by default True.
aux_output_layers : list[int], optional
Layers for auxiliary outputs, when None it defaults to [9, 14, 19].
aux_weights : list[float], optional
Weights for auxiliary outputs, when None it defaults [0.3, 0.3, 0.3].
load_backbone_path : Optional[str], optional
Path to load the backbone model, by default None.
freeze_backbone_on_load : bool, optional
Whether to freeze the backbone model on load, by default True.
learning_rate : float, optional
Learning rate, by default 1e-3.
loss_weights : Optional[list[float]], optional
Weights for the loss function, by default None.
original_resolution : Optional[Tuple[int, int]], optional
The original resolution of the input image in the pre-training
weights. When None, positional embeddings will not be interpolated.
Defaults to None.
head_lr_factor : float, optional
Learning rate factor for the head. used if you need different
learning rates for backbone and prediction head, by default 1.0.
test_engine : Optional[_Engine], optional
Engine used for test and validation steps. When None, behavior of
all steps, training, testing and validation is the same, by default None.
.. py:method:: _compute_metrics(y_hat, y, step_name)
.. py:method:: _loss_func(y_hat, y)
Calculate the loss between the output and the input data.
Parameters
----------
y_hat : torch.Tensor
The output data from the forward pass.
y : torch.Tensor
The input data/label.
Returns
-------
torch.Tensor
The loss value.
.. py:method:: _single_step(batch, batch_idx, step_name)
Perform a single step of the training/validation loop.
Parameters
----------
batch : torch.Tensor
The input data.
batch_idx : int
The index of the batch.
step_name : str
The name of the step, either "train" or "val".
Returns
-------
torch.Tensor
The loss value.
.. py:attribute:: aux_weights
:value: None
.. py:method:: 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.
Return:
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.
.. code-block:: python
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
:class:`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.
.. testcode::
# 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 :class:`~lightning.pytorch.core.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 :class:`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 :meth:`optimizer_step` hook.
.. py:method:: create_from_dict(config)
:staticmethod:
.. py:method:: forward(x)
Same as :meth:`torch.nn.Module.forward`.
Args:
*args: Whatever you decide to pass into the forward method.
**kwargs: Keyword arguments are also possible.
Return:
Your model's output
.. py:attribute:: head_lr_factor
:value: 1.0
.. py:attribute:: learning_rate
:value: 0.001
.. py:method:: load_backbone(path, freeze = False)
.. py:attribute:: loss_fn
:value: None
.. py:attribute:: metrics
.. py:attribute:: model
.. py:attribute:: num_classes
:value: 1000
.. py:attribute:: optimizer_type
:value: None
.. py:method:: predict_step(batch, batch_idx, dataloader_idx = None)
Step function called during :meth:`~lightning.pytorch.trainer.trainer.Trainer.predict`. By default, it calls
:meth:`~lightning.pytorch.core.LightningModule.forward`. Override to add any processing logic.
The :meth:`~lightning.pytorch.core.LightningModule.predict_step` is used
to scale inference on multi-devices.
To prevent an OOM error, it is possible to use :class:`~lightning.pytorch.callbacks.BasePredictionWriter`
callback to write the predictions to disk or database after each batch or on epoch end.
The :class:`~lightning.pytorch.callbacks.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.
Args:
batch: The output of your data iterable, normally a :class:`~torch.utils.data.DataLoader`.
batch_idx: The index of this batch.
dataloader_idx: The index of the dataloader that produced this batch.
(only if multiple dataloaders used)
Return:
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)
.. py:attribute:: test_engine
:value: None
.. py:method:: test_step(batch, batch_idx)
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.
Args:
batch: The output of your data iterable, normally a :class:`~torch.utils.data.DataLoader`.
batch_idx: The index of this batch.
dataloader_idx: The index of the dataloader that produced this batch.
(only if multiple dataloaders used)
Return:
- :class:`~torch.Tensor` - The loss tensor
- ``dict`` - A dictionary. Can include any keys, but must include the key ``'loss'``.
- ``None`` - Skip to the next batch.
.. code-block:: python
# 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, :meth:`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.
.. code-block:: python
# 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 :meth:`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.
.. py:method:: training_step(batch, batch_idx)
Here you compute and return the training loss and some additional metrics for e.g. the progress bar or
logger.
Args:
batch: The output of your data iterable, normally a :class:`~torch.utils.data.DataLoader`.
batch_idx: The index of this batch.
dataloader_idx: The index of the dataloader that produced this batch.
(only if multiple dataloaders used)
Return:
- :class:`~torch.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:
.. code-block:: python
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.
.. py:method:: validation_step(batch, batch_idx)
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.
Args:
batch: The output of your data iterable, normally a :class:`~torch.utils.data.DataLoader`.
batch_idx: The index of this batch.
dataloader_idx: The index of the dataloader that produced this batch.
(only if multiple dataloaders used)
Return:
- :class:`~torch.Tensor` - The loss tensor
- ``dict`` - A dictionary. Can include any keys, but must include the key ``'loss'``.
- ``None`` - Skip to the next batch.
.. code-block:: python
# 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, :meth:`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.
.. code-block:: python
# 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 :meth:`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.
.. py:class:: SimpleSupervisedModel(backbone, fc, loss_fn, adapter = None, learning_rate = 0.001, flatten = True, train_metrics = None, val_metrics = None, test_metrics = None, freeze_backbone = False)
Bases: :py:obj:`lightning.LightningModule`
Simple pipeline for supervised models.
This class implements a very common deep learning pipeline, which is
composed by the following steps:
1. Make a forward pass with the input data on the backbone model;
2. Make a forward pass with the input data on the fc model;
3. Compute the loss between the output and the label data;
4. Optimize the model (backbone and FC) parameters with respect to the loss.
This reduces the code duplication for autoencoder models, and makes it
easier to implement new models by only changing the backbone model. More
complex models, that does not follow this pipeline, should not inherit from
this class.
Note that, for this class the input data is a tuple of tensors, where the
first tensor is the input data and the second tensor is the mask or label.
Initialize the model with the backbone, fc, loss function and
metrics. Metrics are used to evaluate the model during training,
validation, testing or prediction. It will be logged using
lightning logger at the end of each epoch. Metrics should implement
the `torchmetrics.Metric` interface.
Parameters
----------
backbone : torch.nn.Module
The backbone model. Usually the encoder/decoder part of the model.
fc : torch.nn.Module
The fully connected model, usually used to classification tasks.
Use `torch.nn.Identity()` if no FC model is needed.
loss_fn : torch.nn.Module
The function used to compute the loss.
learning_rate : float, optional
The learning rate to Adam optimizer, by default 1e-3
flatten : bool, optional
If `True` the input data will be flattened before passing through
the fc model, by default True
train_metrics : Dict[str, Metric], optional
The metrics to be used during training, by default None
val_metrics : Dict[str, Metric], optional
The metrics to be used during validation, by default None
test_metrics : Dict[str, Metric], optional
The metrics to be used during testing, by default None
predict_metrics : Dict[str, Metric], optional
The metrics to be used during prediction, by default None
.. py:method:: _compute_metrics(y_hat, y, step_name)
Calculate the metrics for the given step.
Parameters
----------
y_hat : torch.Tensor
The output data from the forward pass.
y : torch.Tensor
The input data/label.
step_name : str
Name of the step. It will be used to get the metrics from the
`self.metrics` attribute.
Returns
-------
Dict[str, torch.Tensor]
A dictionary with the metrics values.
.. py:method:: _loss_func(y_hat, y)
Calculate the loss between the output and the input data.
Parameters
----------
y_hat : torch.Tensor
The output data from the forward pass.
y : torch.Tensor
The input data/label.
Returns
-------
torch.Tensor
The loss value.
.. py:method:: _single_step(batch, batch_idx, step_name)
Perform a single train/validation/test step. It consists in making a
forward pass with the input data on the backbone model, computing the
loss between the output and the input data, and logging the loss.
Parameters
----------
batch : torch.Tensor
The input data. It must be a 2-element tuple of tensors, where the
first tensor is the input data and the second tensor is the mask.
batch_idx : int
The index of the batch.
step_name : str
The name of the step. It will be used to log the loss. The possible
values are: "train", "val" and "test". The loss will be logged as
"{step_name}_loss".
Returns
-------
torch.Tensor
A tensor with the loss value.
.. py:attribute:: adapter
:value: None
.. py:attribute:: backbone
.. py:method:: 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.
Return:
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.
.. code-block:: python
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
:class:`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.
.. testcode::
# 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 :class:`~lightning.pytorch.core.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 :class:`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 :meth:`optimizer_step` hook.
.. py:attribute:: fc
.. py:attribute:: flatten
:value: True
.. py:method:: forward(x)
Perform a forward pass with the input data on the backbone model.
Parameters
----------
x : torch.Tensor
The input data.
Returns
-------
torch.Tensor
The output data from the forward pass.
.. py:attribute:: freeze_backbone
:value: False
.. py:attribute:: learning_rate
:value: 0.001
.. py:attribute:: loss_fn
.. py:attribute:: metrics
.. py:method:: predict_step(batch, batch_idx, dataloader_idx=None)
Step function called during :meth:`~lightning.pytorch.trainer.trainer.Trainer.predict`. By default, it calls
:meth:`~lightning.pytorch.core.LightningModule.forward`. Override to add any processing logic.
The :meth:`~lightning.pytorch.core.LightningModule.predict_step` is used
to scale inference on multi-devices.
To prevent an OOM error, it is possible to use :class:`~lightning.pytorch.callbacks.BasePredictionWriter`
callback to write the predictions to disk or database after each batch or on epoch end.
The :class:`~lightning.pytorch.callbacks.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.
Args:
batch: The output of your data iterable, normally a :class:`~torch.utils.data.DataLoader`.
batch_idx: The index of this batch.
dataloader_idx: The index of the dataloader that produced this batch.
(only if multiple dataloaders used)
Return:
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)
.. py:method:: test_step(batch, batch_idx)
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.
Args:
batch: The output of your data iterable, normally a :class:`~torch.utils.data.DataLoader`.
batch_idx: The index of this batch.
dataloader_idx: The index of the dataloader that produced this batch.
(only if multiple dataloaders used)
Return:
- :class:`~torch.Tensor` - The loss tensor
- ``dict`` - A dictionary. Can include any keys, but must include the key ``'loss'``.
- ``None`` - Skip to the next batch.
.. code-block:: python
# 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, :meth:`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.
.. code-block:: python
# 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 :meth:`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.
.. py:method:: training_step(batch, batch_idx)
Here you compute and return the training loss and some additional metrics for e.g. the progress bar or
logger.
Args:
batch: The output of your data iterable, normally a :class:`~torch.utils.data.DataLoader`.
batch_idx: The index of this batch.
dataloader_idx: The index of the dataloader that produced this batch.
(only if multiple dataloaders used)
Return:
- :class:`~torch.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:
.. code-block:: python
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.
.. py:method:: validation_step(batch, batch_idx)
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.
Args:
batch: The output of your data iterable, normally a :class:`~torch.utils.data.DataLoader`.
batch_idx: The index of this batch.
dataloader_idx: The index of the dataloader that produced this batch.
(only if multiple dataloaders used)
Return:
- :class:`~torch.Tensor` - The loss tensor
- ``dict`` - A dictionary. Can include any keys, but must include the key ``'loss'``.
- ``None`` - Skip to the next batch.
.. code-block:: python
# 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, :meth:`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.
.. code-block:: python
# 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 :meth:`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.
.. py:class:: UNet(n_channels = 1, bilinear = False, learning_rate = 0.001, loss_fn = None, **kwargs)
Bases: :py:obj:`minerva.models.nets.base.SimpleSupervisedModel`
This class is a simple implementation of the U-Net model, which is a
convolutional neural network used for image segmentation. The model consists
of a contracting path (encoder) and an expansive path (decoder). The
contracting path follows the typical architecture of a convolutional neural
network, with repeated applications of convolutions and max pooling layers.
The expansive path consists of up-convolutions and concatenation of feature
maps from the contracting path. The model also has skip connections, which
allows the expansive path to use information from the contracting path at
multiple resolutions. The U-Net model was originally proposed by
Ronneberger, Fischer, and Brox in 2015.
This architecture, handles arbitrary input sizes, and returns an output of
the same size as the input. The expected input size is (N, C, H, W), where N
is the batch size, C is the number of channels, H is the height of the input
image, and W is the width of the input image.
Note that, for this implementation, the input batch is a single tensor and
not a tuple of tensors (e.g., data and label).
Note that this class wrappers the `_UNet` class, which is the actual
implementation of the U-Net model, into a `SimpleReconstructionNet` class,
which is a simple autoencoder pipeline for reconstruction tasks.
References
----------
Ronneberger, Olaf, Philipp Fischer, and Thomas Brox. "U-net: Convolutional
networks for biomedical image segmentation." Medical Image Computing and
Computer-Assisted Intervention-MICCAI 2015: 18th International Conference,
Munich, Germany, October 5-9, 2015, Proceedings, Part III 18. Springer
International Publishing, 2015.
Wrapper implementation of the U-Net model.
Parameters
----------
n_channels : int, optional
The number of channels of the input, by default 1
bilinear : bool, optional
If `True` use bilinear interpolation for upsampling, by default
False.
learning_rate : float, optional
The learning rate to Adam optimizer, by default 1e-3
loss_fn : torch.nn.Module, optional
The function used to compute the loss. If `None`, it will be used
the MSELoss, by default None.
kwargs : Dict
Additional arguments to be passed to the `SimpleSupervisedModel`
class.
.. py:class:: WiseNet(in_channels = 1, out_channels = 1, loss_fn = None, learning_rate = 0.001, **kwargs)
Bases: :py:obj:`minerva.models.nets.base.SimpleSupervisedModel`
Simple pipeline for supervised models.
This class implements a very common deep learning pipeline, which is
composed by the following steps:
1. Make a forward pass with the input data on the backbone model;
2. Make a forward pass with the input data on the fc model;
3. Compute the loss between the output and the label data;
4. Optimize the model (backbone and FC) parameters with respect to the loss.
This reduces the code duplication for autoencoder models, and makes it
easier to implement new models by only changing the backbone model. More
complex models, that does not follow this pipeline, should not inherit from
this class.
Note that, for this class the input data is a tuple of tensors, where the
first tensor is the input data and the second tensor is the mask or label.
Initialize the model with the backbone, fc, loss function and
metrics. Metrics are used to evaluate the model during training,
validation, testing or prediction. It will be logged using
lightning logger at the end of each epoch. Metrics should implement
the `torchmetrics.Metric` interface.
Parameters
----------
backbone : torch.nn.Module
The backbone model. Usually the encoder/decoder part of the model.
fc : torch.nn.Module
The fully connected model, usually used to classification tasks.
Use `torch.nn.Identity()` if no FC model is needed.
loss_fn : torch.nn.Module
The function used to compute the loss.
learning_rate : float, optional
The learning rate to Adam optimizer, by default 1e-3
flatten : bool, optional
If `True` the input data will be flattened before passing through
the fc model, by default True
train_metrics : Dict[str, Metric], optional
The metrics to be used during training, by default None
val_metrics : Dict[str, Metric], optional
The metrics to be used during validation, by default None
test_metrics : Dict[str, Metric], optional
The metrics to be used during testing, by default None
predict_metrics : Dict[str, Metric], optional
The metrics to be used during prediction, by default None
.. py:method:: _single_step(batch, batch_idx, step_name)
Perform a single train/validation/test step. It consists in making a
forward pass with the input data on the backbone model, computing the
loss between the output and the input data, and logging the loss.
Parameters
----------
batch : torch.Tensor
The input data. It must be a 2-element tuple of tensors, where the
first tensor is the input data and the second tensor is the mask.
batch_idx : int
The index of the batch.
step_name : str
The name of the step. It will be used to log the loss. The possible
values are: "train", "val" and "test". The loss will be logged as
"{step_name}_loss".
Returns
-------
torch.Tensor
A tensor with the loss value.
.. py:method:: predict_step(batch, batch_idx, dataloader_idx=None)
Step function called during :meth:`~lightning.pytorch.trainer.trainer.Trainer.predict`. By default, it calls
:meth:`~lightning.pytorch.core.LightningModule.forward`. Override to add any processing logic.
The :meth:`~lightning.pytorch.core.LightningModule.predict_step` is used
to scale inference on multi-devices.
To prevent an OOM error, it is possible to use :class:`~lightning.pytorch.callbacks.BasePredictionWriter`
callback to write the predictions to disk or database after each batch or on epoch end.
The :class:`~lightning.pytorch.callbacks.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.
Args:
batch: The output of your data iterable, normally a :class:`~torch.utils.data.DataLoader`.
batch_idx: The index of this batch.
dataloader_idx: The index of the dataloader that produced this batch.
(only if multiple dataloaders used)
Return:
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)