SDK ML Classification API

This module contains functions to compute and display classification KPIs

metavision_ml.classification.utils_metrics.calculate_time_to_prediction(pred_seq, label_seq, delta_t=10000)

calculate KPI: time to 1st correct prediction & draw histogram of the time to prediction statistics

Parameters

  • pred_seq (list) – list of prediction tensors for each data sample

  • labels_all (list) – list of label tensors for each data sample

  • delta_t (int) – time interval of data sample

metavision_ml.classification.utils_metrics.evaluate_preds_seq(preds_seq, labels_seq, res_per_recording, hparams, latency_seq)

Inspect the test result by plotting the recording image together with prediction sequence

Parameters

  • preds_seq (list) – nested list of prediction sequences

  • labels_seq (list) – nested list of labeling sequences

  • res_per_recording (defaultdict) – defaultdict of time stamp, prediction and label vectors per HDF5 file

  • hparams (dict) – hyperparameters

  • latency_seq – list of time to prediction for each data sample

metavision_ml.classification.utils_metrics.get_1st_nonzeros(tensor)

Get the 1st nonzero item along the last axis of the tensor If tensor only contains zeros, get the last item index :param tensor: input tensor :type tensor: torch.Tensor

metavision_ml.classification.utils_metrics.plot_cm(preds_all, labels_all, labels)

Plot confusion metrics & error map by masking the diagonal values

Parameters

  • preds_all (torch.Tensor) – predictions

  • labels_all (torch.Tensor) – GT

  • labels (list) – list of all class labels

metavision_ml.classification.utils_metrics.plot_precision_recall_curve(preds_all, labels_all, labels)

plot the PR-curve

Parameters

  • preds_all (torch.Tensor) – predictions

  • labels_all (torch.Tensor) – GT

  • labels (list) – list of all class labels

metavision_ml.classification.utils_metrics.plot_roc(preds_all, labels_all, labels)

plot roc curve with auc_roc score

Parameters

  • preds_all (torch.Tensor) – predictions

  • labels_all (torch.Tensor) – GT

  • labels (list) – list of all class labels

metavision_ml.classification.utils_metrics.unpack_metrics_dict(metrics_per_category, label)

Unpack the dense metrics

Parameters

  • metrics_per_category (dict) – the torchmetrics result calculated per category

  • label (list) – list of class labels

This data module is a wrapper around SequentialDataLoader for the classification module.

class metavision_ml.classification.data_module.ClassificationDataModule(hparams, data_dir: str = '')

Data Module for classification Applies some data augmentation on top.

prepare_data_per_node

If True, each LOCAL_RANK=0 will call prepare data. Otherwise only NODE_RANK=0, LOCAL_RANK=0 will prepare data.

allow_zero_length_dataloader_with_multiple_devices

If True, dataloader with zero length within local rank is allowed. Default value is False.

test_dataloader()

An iterable or collection of iterables specifying test samples.

For more information about multiple dataloaders, see this section.

For data processing use the following pattern:

  • download in prepare_data()

  • process and split in setup()

However, the above are only necessary for distributed processing.

Warning

do not assign state in prepare_data

  • test()

  • prepare_data()

  • setup()

Note

Lightning tries to add the correct sampler for distributed and arbitrary hardware. There is no need to set it yourself.

Note

If you don’t need a test dataset and a test_step(), you don’t need to implement this method.

train_dataloader()

An iterable or collection of iterables specifying training samples.

For more information about multiple dataloaders, see this section.

The dataloader you return will not be reloaded unless you set :paramref:`~pytorch_lightning.trainer.trainer.Trainer.reload_dataloaders_every_n_epochs` to a positive integer.

For data processing use the following pattern:

  • download in prepare_data()

  • process and split in setup()

However, the above are only necessary for distributed processing.

Warning

do not assign state in prepare_data

  • fit()

  • prepare_data()

  • setup()

Note

Lightning tries to add the correct sampler for distributed and arbitrary hardware. There is no need to set it yourself.

transfer_batch_to_device(batch, device, dataloader_idx)

Override this hook if your DataLoader returns tensors wrapped in a custom data structure.

The data types listed below (and any arbitrary nesting of them) are supported out of the box:

  • torch.Tensor or anything that implements .to(…)

  • list

  • dict

  • tuple

For anything else, you need to define how the data is moved to the target device (CPU, GPU, TPU, …).

Note

This hook should only transfer the data and not modify it, nor should it move the data to any other device than the one passed in as argument (unless you know what you are doing). To check the current state of execution of this hook you can use self.trainer.training/testing/validating/predicting so that you can add different logic as per your requirement.

Parameters

  • batch – A batch of data that needs to be transferred to a new device.

  • device – The target device as defined in PyTorch.

  • dataloader_idx – The index of the dataloader to which the batch belongs.

Returns

A reference to the data on the new device.

Example:

def transfer_batch_to_device(self, batch, device, dataloader_idx):
    if isinstance(batch, CustomBatch):
        # move all tensors in your custom data structure to the device
        batch.samples = batch.samples.to(device)
        batch.targets = batch.targets.to(device)
    elif dataloader_idx == 0:
        # skip device transfer for the first dataloader or anything you wish
        pass
    else:
        batch = super().transfer_batch_to_device(batch, device, dataloader_idx)
    return batch
Raises

MisconfigurationException – If using IPUs, Trainer(accelerator='ipu').

See also

  • move_data_to_device()

  • apply_to_collection()

val_dataloader()

An iterable or collection of iterables specifying validation samples.

For more information about multiple dataloaders, see this section.

The dataloader you return will not be reloaded unless you set :paramref:`~pytorch_lightning.trainer.trainer.Trainer.reload_dataloaders_every_n_epochs` to a positive integer.

It’s recommended that all data downloads and preparation happen in prepare_data().

  • fit()

  • validate()

  • prepare_data()

  • setup()

Note

Lightning tries to add the correct sampler for distributed and arbitrary hardware There is no need to set it yourself.

Note

If you don’t need a validation dataset and a validation_step(), you don’t need to implement this method.

metavision_ml.classification.data_module.load_classes(metadata, batch_start_time, duration, tensor, **kwargs)

Function to fetch boxes and preprocess them. Should be passed to a SequentialDataLoader.

Examples

>>> from functools import partial
>>> n_classes = 21
>>> class_lookup = np.arange(n_classes)  # each class is mapped to itself
>>> load_boxes_function = partial(load_boxes, class_lookup=class_lookup)
Parameters

  • metadata (object) – Record details.

  • batch_start_time (int) – (us) Where to seek in the file to load corresponding bounding boxes

  • duration (int) – (us) How long to load events from bounding box file

  • tensor (np.ndarray) – Current preprocessed input, can be used for data dependent preprocessing, for instance remove boxes without any features in them.

  • class_lookup (np.array) – Look up array for class indices.

  • labelling_delta_t (int) – Indicates the period of labelling in order to only consider time bins with actual labels when computing the loss.

Returns

List of structured array of dtype EventBbox corresponding to each time bins. frame_is_labeled (np.ndarray): This boolean mask array of length num_tbins indicates whether the frame contains a label. It is used to differentiate between time bins that actually contain an empty label (for instance no bounding boxes) from time bins that weren’t labeled due to cost constraints. The latter time bins shouldn’t contribute to supervised losses used during training.

Return type

labels (List[np.ndarray])

Models for classification

class metavision_ml.classification.models.ConvRNNClassifier(cin=1, base=16, cout=256, num_classes=2)

ConvRNN Classifier

Feed-Forward + RNN light model

Parameters

  • cin (int) – aaa

  • base (int) – bbb

  • cout (int) – ccc

  • num_classes (int) – ddd

Initialize internal Module state, shared by both nn.Module and ScriptModule.

forward(x)

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 metavision_ml.classification.models.LeNetClassifier(cin=1, base=6, cout=256, num_classes=2)

LeNet RNN

Initialize internal Module state, shared by both nn.Module and ScriptModule.

forward(x)

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 metavision_ml.classification.models.Mobilenetv2Classifier(cin=2, width_mul=1.0, num_classes=2, round_nearest=8, **kwargs)

Mobilenetv2

Modified Feed-Forward architecture

Initialize internal Module state, shared by both nn.Module and ScriptModule.

forward(x)

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 metavision_ml.classification.models.SqueezenetClassifier(cin=2, num_classes=2, **kwargs)

Mobilenetv2

Modified Feed-Forward architecture

Initialize internal Module state, shared by both nn.Module and ScriptModule.

forward(x)

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.

Pytorch Lightning Module for training a classifier

class metavision_ml.classification.lightning_model.ClassificationModel(hparams: argparse.Namespace)

Pytorch Lightning model for neural network to predict class of scene.

Parameters

hparams (argparse.Namespace) – argparse from train.py application

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.

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.

demo_video(test_data, epoch=0, num_batches=100, show_video=False, show_pred=True, fps=30)

This runs our classifier on several videos of the test dataset

Parameters

  • test_data (object) – Dataloader

  • epoch (int, optional) – Index of the epoch. Used to name the video

  • num_batches (int, optional) – Number of batches used to create the video

  • show_video (boolean, optional) – Whether to display the demo

  • show_pred (boolean, optional) – Whether to show the prediction results as well. Set it to “False” to only inspect the input data

  • fps (int, optional) – Video output frame rate

forward(batch)

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

load_pretrained(checkpoint_path)

Loads a pretrained detector (of this class) and transfer the weights to this module for fine-tuning.

In addition, it may remap the old classification weights if some overlap exists between old and new list of classes.

Parameters

checkpoint_path (str) – path to checkpoint of pretrained detector.

on_test_epoch_end()

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

on_test_epoch_start()

Called in the test loop at the very beginning of the epoch.

on_train_epoch_end()

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()
on_train_epoch_start()

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

on_validation_epoch_end()

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

on_validation_epoch_start()

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

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.

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.

training_step(batch, batch_nb)

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)

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.

This data module is a wrapper around NonSequentialDataLoader for the classification module.

class metavision_ml.classification.fnn_data_module.FNNClassificationDataModule(hparams, data_dir: str = '')

FNN Data Module for classification Applies some data augmentation on top.

prepare_data_per_node

If True, each LOCAL_RANK=0 will call prepare data. Otherwise only NODE_RANK=0, LOCAL_RANK=0 will prepare data.

allow_zero_length_dataloader_with_multiple_devices

If True, dataloader with zero length within local rank is allowed. Default value is False.

test_dataloader()

An iterable or collection of iterables specifying test samples.

For more information about multiple dataloaders, see this section.

For data processing use the following pattern:

  • download in prepare_data()

  • process and split in setup()

However, the above are only necessary for distributed processing.

Warning

do not assign state in prepare_data

  • test()

  • prepare_data()

  • setup()

Note

Lightning tries to add the correct sampler for distributed and arbitrary hardware. There is no need to set it yourself.

Note

If you don’t need a test dataset and a test_step(), you don’t need to implement this method.

train_dataloader()

An iterable or collection of iterables specifying training samples.

For more information about multiple dataloaders, see this section.

The dataloader you return will not be reloaded unless you set :paramref:`~pytorch_lightning.trainer.trainer.Trainer.reload_dataloaders_every_n_epochs` to a positive integer.

For data processing use the following pattern:

  • download in prepare_data()

  • process and split in setup()

However, the above are only necessary for distributed processing.

Warning

do not assign state in prepare_data

  • fit()

  • prepare_data()

  • setup()

Note

Lightning tries to add the correct sampler for distributed and arbitrary hardware. There is no need to set it yourself.

val_dataloader()

An iterable or collection of iterables specifying validation samples.

For more information about multiple dataloaders, see this section.

The dataloader you return will not be reloaded unless you set :paramref:`~pytorch_lightning.trainer.trainer.Trainer.reload_dataloaders_every_n_epochs` to a positive integer.

It’s recommended that all data downloads and preparation happen in prepare_data().

  • fit()

  • validate()

  • prepare_data()

  • setup()

Note

Lightning tries to add the correct sampler for distributed and arbitrary hardware There is no need to set it yourself.

Note

If you don’t need a validation dataset and a validation_step(), you don’t need to implement this method.

Pytorch Lightning Module for training a fnn classifier

class metavision_ml.classification.fnn_lightning_model.FNNClassificationModel(hparams: argparse.Namespace)

Pytorch Lightning model for neural network to predict class of scene.

Parameters

hparams (argparse.Namespace) – argparse from train.py application

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.

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.

demo_video(test_data, epoch=0, num_batches=100, show_video=False, show_pred=True, fps=30)

This runs our classifier on several videos of the test dataset

Parameters

  • test_data (object) – Dataloader

  • epoch (int, optional) – Index of the epoch. Used to name the video

  • num_batches (int, optional) – Number of batches used to create the video

  • show_video (boolean, optional) – Whether to display the demo

  • show_pred (boolean, optional) – Whether to show the prediction results as well. Set it to “False” to only inspect the input data

  • fps (int, optional) – Video output frame rate

forward(input)

Input in the dimension of (batch_size, channels, height, width) Output in the dimension of (batch_size, num_classes)

load_pretrained(checkpoint_path)

Loads a pretrained detector (of this class) and transfer the weights to this module for fine-tuning.

Parameters

checkpoint_path (str) – path to checkpoint of pretrained detector.

on_test_epoch_end()

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

on_test_epoch_start()

Called in the test loop at the very beginning of the epoch.

on_train_epoch_end()

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()
on_train_epoch_start()

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

on_validation_epoch_end()

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

on_validation_epoch_start()

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

preprocess_inputs(batch)
Input:

event_frames, labels = batch event_frames in the shape of (batch_size, num_ev_reps, channels, height, width) labels in the shape of (batch_size, num_ev_reps)

Output:

we use the last label to represent the label in the group of num_ev_reps event frames event_frames in the shape of (batch_size, num_ev_reps * channels, height, width) labels in the shape of (batch_size)

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.

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.

training_step(batch, batch_nb)

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)

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.