Hooks¶

class classy_vision.hooks.CheckpointHook(checkpoint_folder: str, input_args: Optional[Any] = None, phase_types: Optional[Collection[str]] = None, checkpoint_period: int = 1)¶

Hook to checkpoint a model’s task.

Saves the checkpoints in checkpoint_folder.

__init__(checkpoint_folder: str, input_args: Optional[Any] = None, phase_types: Optional[Collection[str]] = None, checkpoint_period: int = 1) → None¶

The constructor method of CheckpointHook.

Parameters

checkpoint_folder – Folder to store checkpoints in
input_args – Any arguments to save about the runtime setup. For example, it is useful to store the config that was used to instantiate the model.
phase_types – If phase_types is specified, only checkpoint on those phase types. Each item in phase_types must be either “train” or “test”. If not specified, it is set to checkpoint after “train” phases.
checkpoint_period – Checkpoint at the end of every x phases (default 1)

on_end(*args, **kwargs) → None¶

Derived classes can set their hook functions to this.

This is useful if they want those hook functions to not do anything.

on_phase_end(task) → None¶

Checkpoint the task every checkpoint_period phases.

We do not necessarily checkpoint the task at the end of every phase.

on_phase_start(*args, **kwargs) → None¶

Derived classes can set their hook functions to this.

This is useful if they want those hook functions to not do anything.

on_start(task) → None¶: Called at the start of training.

on_step(*args, **kwargs) → None¶

Derived classes can set their hook functions to this.

This is useful if they want those hook functions to not do anything.

class classy_vision.hooks.ClassyHook¶

Base class for hooks.

Hooks allow to inject behavior at different places of the training loop, which are listed below in the chronological order.

on_start -> on_phase_start ->
on_step -> on_phase_end -> on_end

Deriving classes should call super().__init__() and store any state in self.state. Any state added to this property should be serializable. E.g. -

class MyHook(ClassyHook):
    def __init__(self, a, b):
        super().__init__()
        self.state.a = [1,2,3]
        self.state.b = "my_hook"
        # the following line is not allowed
        # self.state.my_lambda = lambda x: x^2

__init__()¶

get_classy_state() → Dict[str, Any]¶

Get the state of the ClassyHook.

The returned state is used for checkpointing.

Returns: A state dictionary containing the state of the hook.

classmethod name() → str¶: Returns the name of the class.

abstract on_end(task) → None¶: Called at the end of training.

abstract on_phase_end(task) → None¶: Called at the end of each phase (epoch).

abstract on_phase_start(task) → None¶: Called at the start of each phase.

abstract on_start(task) → None¶: Called at the start of training.

abstract on_step(task) → None¶: Called each time after parameters have been updated by the optimizer.

set_classy_state(state_dict: Dict[str, Any]) → None¶

Set the state of the ClassyHook.

Parameters: state_dict – The state dictionary. Must be the output of a call to get_classy_state().

This is used to load the state of the hook from a checkpoint.

class classy_vision.hooks.ClassyHookFunctions(value)¶: Enumeration of all the hook functions in the ClassyHook class.

class classy_vision.hooks.ExponentialMovingAverageModelHook(decay: float, consider_bn_buffers: bool = True, device: str = 'gpu')¶

Hook which keeps a track of the exponential moving average (EMA) of the model’s parameters and applies the EMA params to the model during the test phases.

Saving the state in cpu will save gpu memory, but will make training slower since the model parameters will need to be moved to cpu before the averaging.

Note

This hooks stores two additional copies of the model’s parameters, which will increase memory usage significantly.

__init__(decay: float, consider_bn_buffers: bool = True, device: str = 'gpu') → None¶

The constructor method of ExponentialMovingAverageModelHook.

Parameters

decay – EMA decay factor, should be in [0, 1]. A decay of 0 corresponds to always using the latest value (no EMA) and a decay of 1 corresponds to not updating weights after initialization.
consider_bn_buffers – Whether to apply EMA to batch norm buffers
device – Device to store the model state.

get_model_state_iterator(model: torch.nn.modules.module.Module) → Iterable[Tuple[str, Any]]¶: Get an iterator over the model state to apply EMA to.

on_end(*args, **kwargs) → None¶

Derived classes can set their hook functions to this.

This is useful if they want those hook functions to not do anything.

on_phase_end(task) → None¶: Called at the end of each phase (epoch).

on_phase_start(task) → None¶: Called at the start of each phase.

on_start(task) → None¶: Called at the start of training.

on_step(task) → None¶: Called each time after parameters have been updated by the optimizer.

set_model_state(task, use_ema: bool) → None¶: Depending on use_ema, set the appropriate state for the model.

class classy_vision.hooks.LossLrMeterLoggingHook(log_freq: Optional[int] = None)¶

Logs the loss, optimizer LR, and meters. Logs at the end of a phase.

__init__(log_freq: Optional[int] = None) → None¶

The constructor method of LossLrMeterLoggingHook.

Parameters: log_freq – if specified, also logs every log_freq batches.

on_end(*args, **kwargs) → None¶

Derived classes can set their hook functions to this.

This is useful if they want those hook functions to not do anything.

on_phase_end(task) → None¶: Log the loss, optimizer LR, and meters for the phase.

on_phase_start(*args, **kwargs) → None¶

Derived classes can set their hook functions to this.

This is useful if they want those hook functions to not do anything.

on_start(task) → None¶: Called at the start of training.

on_step(task) → None¶: Log the LR every log_freq batches, if log_freq is not None.

class classy_vision.hooks.ModelComplexityHook¶

Logs the number of paramaters and forward pass FLOPs and activations of the model.

__init__() → None¶

on_end(*args, **kwargs) → None¶

Derived classes can set their hook functions to this.

This is useful if they want those hook functions to not do anything.

on_phase_end(*args, **kwargs) → None¶

Derived classes can set their hook functions to this.

This is useful if they want those hook functions to not do anything.

on_phase_start(*args, **kwargs) → None¶

Derived classes can set their hook functions to this.

This is useful if they want those hook functions to not do anything.

on_start(task) → None¶: Measure number of parameters, FLOPs and activations.

on_step(*args, **kwargs) → None¶

Derived classes can set their hook functions to this.

This is useful if they want those hook functions to not do anything.

class classy_vision.hooks.ModelTensorboardHook(tb_writer)¶

Shows the model graph in TensorBoard <https ://www.tensorflow.org/tensorboard>_.

__init__(tb_writer) → None¶

The constructor method of ModelTensorboardHook.

Parameters

tb_writer – `Tensorboard SummaryWriter <https://tensorboardx.
readthedocs.io/en/latest/tensorboard.html#tensorboardX. –
None (SummaryWriter>`_ instance or) –

classmethod from_config(config: [Dict[str, Any]]) → classy_vision.hooks.model_tensorboard_hook.ModelTensorboardHook¶: The config is expected to include the key “summary_writer” with arguments which correspond to those listed at <https://tensorboardx. readthedocs.io/en/latest/tensorboard.html#tensorboardX.SummaryWriter>:

on_end(*args, **kwargs) → None¶

Derived classes can set their hook functions to this.

This is useful if they want those hook functions to not do anything.

on_phase_end(*args, **kwargs) → None¶

Derived classes can set their hook functions to this.

This is useful if they want those hook functions to not do anything.

on_phase_start(*args, **kwargs) → None¶

Derived classes can set their hook functions to this.

This is useful if they want those hook functions to not do anything.

on_start(task) → None¶: Plot the model on Tensorboard.

on_step(*args, **kwargs) → None¶

Derived classes can set their hook functions to this.

This is useful if they want those hook functions to not do anything.

class classy_vision.hooks.OutputCSVHook(folder, id_key='id', delimiter='\t')¶

__init__(folder, id_key='id', delimiter='\t') → None¶

on_end(task) → None¶: Called at the end of training.

on_phase_end(task) → None¶: Called at the end of each phase (epoch).

on_phase_start(*args, **kwargs) → None¶

Derived classes can set their hook functions to this.

This is useful if they want those hook functions to not do anything.

on_start(task) → None¶: Called at the start of training.

on_step(task) → None¶

Saves the output of the model to a CSV file.

This hook assumes the dataset provides an “id” key. It also expects the task to provide an output of shape (B, C) where B is the batch size and C is the number of classes. Targets can be either one-hot encoded or single numbers.

class classy_vision.hooks.PreciseBatchNormHook(num_samples: int, cache_samples: bool = False)¶

Hook to compute precise batch normalization statistics.

Batch norm stats are calculated and updated during training, when the weights are also changing, which makes the calculations imprecise. This hook recomputes the batch norm stats at the end of a train phase to make them more precise. See fvcore’s documentation for more information.

__init__(num_samples: int, cache_samples: bool = False) → None¶

The constructor method of PreciseBatchNormHook.

Caches the required number of samples on the CPU during train phases

Parameters

num_samples – Number of samples to calculate the batch norm stats per replica
cache_samples – If True, we cache samples at training stage. This avoids re-creating data loaders, but consumes more memory. If False, we re-create data loader at the end of phase, which might be slow for large dataset, but saves memory.

on_end(*args, **kwargs) → None¶

Derived classes can set their hook functions to this.

This is useful if they want those hook functions to not do anything.

on_phase_end(task) → None¶: Called at the end of each phase (epoch).

on_phase_start(task) → None¶: Called at the start of each phase.

on_start(task) → None¶: Called at the start of training.

on_step(task) → None¶: Called each time after parameters have been updated by the optimizer.

class classy_vision.hooks.ProfilerHook¶

Hook to profile a model and to show model runtime information, such as: the time breakdown in milliseconds of forward/backward pass.

on_end(*args, **kwargs) → None¶

Derived classes can set their hook functions to this.

This is useful if they want those hook functions to not do anything.

on_phase_end(*args, **kwargs) → None¶

Derived classes can set their hook functions to this.

This is useful if they want those hook functions to not do anything.

on_phase_start(*args, **kwargs) → None¶

Derived classes can set their hook functions to this.

This is useful if they want those hook functions to not do anything.

on_start(task) → None¶: Profile the forward pass.

on_step(*args, **kwargs) → None¶

Derived classes can set their hook functions to this.

This is useful if they want those hook functions to not do anything.

class classy_vision.hooks.ProgressBarHook¶

Displays a progress bar to show progress in processing batches.

__init__() → None¶: The constructor method of ProgressBarHook.

on_end(*args, **kwargs) → None¶

Derived classes can set their hook functions to this.

This is useful if they want those hook functions to not do anything.

on_phase_end(task) → None¶: Clear the progress bar at the end of the phase.

on_phase_start(task) → None¶: Create and display a progress bar with 0 progress.

on_start(*args, **kwargs) → None¶

Derived classes can set their hook functions to this.

This is useful if they want those hook functions to not do anything.

on_step(task) → None¶: Update the progress bar with the batch size.

class classy_vision.hooks.TensorboardPlotHook(tb_writer, log_period: int = 10)¶

Hook for writing the losses, learning rates and meters to tensorboard <https ://www.tensorflow.org/tensorboard>_.

Global steps are counted in terms of the number of samples processed.

__init__(tb_writer, log_period: int = 10) → None¶

The constructor method of TensorboardPlotHook.

Parameters

tb_writer – `Tensorboard SummaryWriter <https://tensorboardx.
readthedocs.io/en/latest/tensorboard.html#tensorboardX. –
None (SummaryWriter>`_ instance or) –

classmethod from_config(config: Dict[str, Any]) → classy_vision.hooks.tensorboard_plot_hook.TensorboardPlotHook¶: The config is expected to include the key “summary_writer” with arguments which correspond to those listed at <https://tensorboardx. readthedocs.io/en/latest/tensorboard.html#tensorboardX.SummaryWriter>:

on_end(*args, **kwargs) → None¶

Derived classes can set their hook functions to this.

This is useful if they want those hook functions to not do anything.

on_phase_end(task) → None¶: Add the losses and learning rates to tensorboard.

on_phase_start(task) → None¶: Initialize losses and learning_rates.

on_start(task) → None¶: Called at the start of training.

on_step(task) → None¶: Store the observed learning rates.

class classy_vision.hooks.TorchscriptHook(torchscript_folder: str, use_trace: bool = True, trace_strict: bool = True, device: str = 'cpu')¶

Hook to convert a task model into torch script.

Saves the torch scripts in torchscript_folder.

__init__(torchscript_folder: str, use_trace: bool = True, trace_strict: bool = True, device: str = 'cpu') → None¶

The constructor method of TorchscriptHook.

Parameters

torchscript_folder – Folder to store torch scripts in.
use_trace – set to true for tracing and false for scripting,
trace_strict – run the tracer in a strict mode or not (default: True). Only turn this off when you want the tracer to record your mutable container types (currently list/dict) and you are sure that the container you are using in your problem is a constant structure and does not get used as control flow (if, for) conditions.
device – move to device before saving.

on_end(task) → None¶: Save model into torchscript by the end of training/testing.

on_phase_end(*args, **kwargs) → None¶

Derived classes can set their hook functions to this.

This is useful if they want those hook functions to not do anything.

on_phase_start(*args, **kwargs) → None¶

Derived classes can set their hook functions to this.

This is useful if they want those hook functions to not do anything.

on_start(task) → None¶: Called at the start of training.

on_step(*args, **kwargs) → None¶

Derived classes can set their hook functions to this.

This is useful if they want those hook functions to not do anything.

class classy_vision.hooks.VisdomHook(server: str, port: str, env: str = 'main', title_suffix: str = '')¶

Plots metrics on to Visdom.

Visdom is a flexible tool for creating, organizing, and sharing visualizations: of live, rich data. It supports Python.

__init__(server: str, port: str, env: str = 'main', title_suffix: str = '') → None¶

Parameters

server – host name of the visdom server
port – port of visdom server, such as 8097
env – environment of visdom
title_suffix – suffix that will be appended to the title

on_end(*args, **kwargs) → None¶

Derived classes can set their hook functions to this.

This is useful if they want those hook functions to not do anything.

on_phase_end(task) → None¶: Plot the metrics on visdom.

on_phase_start(*args, **kwargs) → None¶

Derived classes can set their hook functions to this.

This is useful if they want those hook functions to not do anything.

on_start(*args, **kwargs) → None¶

Derived classes can set their hook functions to this.

This is useful if they want those hook functions to not do anything.

on_step(*args, **kwargs) → None¶

Derived classes can set their hook functions to this.

This is useful if they want those hook functions to not do anything.

classy_vision.hooks.build_hook(hook_config: Dict[str, Any])¶

Builds a ClassyHook from a config.

This assumes a ‘name’ key in the config which is used to determine what hook class to instantiate. For instance, a config {“name”: “my_hook”, “foo”: “bar”} will find a class that was registered as “my_hook” (see register_hook()) and call .from_config on it.

classy_vision.hooks.register_hook(name, bypass_checks=False)¶

Registers a ClassyHook subclass.

This decorator allows Classy Vision to instantiate a subclass of ClassyHook from a configuration file, even if the class itself is not part of the base Classy Vision framework. To use it, apply this decorator to a ClassyHook subclass, like this:

@register_hook('custom_hook')
class CustomHook(ClassyHook):
   ...

To instantiate a hook from a configuration file, see build_hook().

Classy Vision

Hooks¶

Classy Vision

Navigation

Related Topics