morpheus.models.dfencoder.autoencoder.AutoEncoder

class AutoEncoder(*, encoder_layers=None, decoder_layers=None, encoder_dropout=None, decoder_dropout=None, encoder_activations=None, decoder_activations=None, activation='relu', min_cats=10, swap_p=0.15, lr=0.01, batch_size=256, eval_batch_size=1024, optimizer='adam', amsgrad=False, momentum=0, betas=(0.9, 0.999), dampening=0, weight_decay=0, lr_decay=None, nesterov=False, verbose=False, device=None, distributed_training=False, logger='basic', logdir='logdir/', project_embeddings=True, run=None, progress_bar=True, n_megabatches=1, scaler='standard', patience=5, preset_cats=None, preset_numerical_scaler_params=None, binary_feature_list=None, loss_scaler='standard', **kwargs)[source]

Bases: torch.nn.modules.module.Module

Methods

add_module(name, module)

Adds a child module to the current module.

apply(fn)

Applies fn recursively to every submodule (as returned by .children()) as well as self.

bfloat16()

Casts all floating point parameters and buffers to bfloat16 datatype.

buffers([recurse])

Returns an iterator over module buffers.

children()

Returns an iterator over immediate children modules.

compute_baseline_performance(in_, out_)

Baseline performance is computed by generating a strong

compute_loss_from_targets(num, bin, cat, ...)

Computes the loss from targets.

cpu()

Moves all model parameters and buffers to the CPU.

cuda([device])

Moves all model parameters and buffers to the GPU.

decode_outputs_to_df(num, bin, cat)

Converts the model outputs of the numerical, binary, and categorical features back into a pandas dataframe.

df_predict(df)

Runs end-to-end model. Interprets output and creates a dataframe. Outputs dataframe with same shape as input containing model predictions.

double()

Casts all floating point parameters and buffers to double datatype.

encode_input(df)

Handles raw df inputs.

eval()

Sets the module in evaluation mode.

extra_repr()

Set the extra representation of the module

fit(train_data[, epochs, val_data, ...])

Does training in the specified mode (indicated by self.distrivuted_training).

float()

Casts all floating point parameters and buffers to float datatype.

forward(*input)

Defines the computation performed at every call.

get_anomaly_score(df)

Returns a per-row loss of the input dataframe.

get_anomaly_score_losses(df)

Run the input dataframe df through the autoencoder to get the recovery losses by feature type (numerical/boolean/categorical).

get_buffer(target)

Returns the buffer given by target if it exists, otherwise throws an error.

get_deep_stack_features(df)

records and outputs all internal representations of input df as row-wise vectors.

get_extra_state()

Returns any extra state to include in the module's state_dict.

get_parameter(target)

Returns the parameter given by target if it exists, otherwise throws an error.

get_representation(df[, layer])

Computes latent feature vector from hidden layer

get_results_from_dataset(dataset, preloaded_df)

Returns a pandas dataframe of inference results and losses for a given dataset.

get_submodule(target)

Returns the submodule given by target if it exists, otherwise throws an error.

half()

Casts all floating point parameters and buffers to half datatype.

ipu([device])

Moves all model parameters and buffers to the IPU.

load_state_dict(state_dict[, strict])

Copies parameters and buffers from state_dict into this module and its descendants.

modules()

Returns an iterator over all modules in the network.

named_buffers([prefix, recurse])

Returns an iterator over module buffers, yielding both the name of the buffer as well as the buffer itself.

named_children()

Returns an iterator over immediate children modules, yielding both the name of the module as well as the module itself.

named_modules([memo, prefix, remove_duplicate])

Returns an iterator over all modules in the network, yielding both the name of the module as well as the module itself.

named_parameters([prefix, recurse])

Returns an iterator over module parameters, yielding both the name of the parameter as well as the parameter itself.

parameters([recurse])

Returns an iterator over module parameters.

prepare_df(df)

Does data preparation on copy of input dataframe.

preprocess_data(df, shuffle_rows_in_batch, ...)

Preprocesses a pandas dataframe df for input into the autoencoder model.

preprocess_train_data(df[, ...])

Wrapper function round self.preprocess_data feeding in the args suitable for a training set.

preprocess_validation_data(df[, ...])

Wrapper function round self.preprocess_data feeding in the args suitable for a validation set.

register_backward_hook(hook)

Registers a backward hook on the module.

register_buffer(name, tensor[, persistent])

Adds a buffer to the module.

register_forward_hook(hook)

Registers a forward hook on the module.

register_forward_pre_hook(hook)

Registers a forward pre-hook on the module.

register_full_backward_hook(hook)

Registers a backward hook on the module.

register_load_state_dict_post_hook(hook)

Registers a post hook to be run after module's load_state_dict is called.

register_module(name, module)

Alias for add_module().

register_parameter(name, param)

Adds a parameter to the module.

requires_grad_([requires_grad])

Change if autograd should record operations on parameters in this module.

set_extra_state(state)

This function is called from load_state_dict() to handle any extra state found within the state_dict.

share_memory()

See torch.Tensor.share_memory_()

state_dict(*args[, destination, prefix, ...])

Returns a dictionary containing references to the whole state of the module.

to(*args, **kwargs)

Moves and/or casts the parameters and buffers.

to_empty(*, device)

Moves the parameters and buffers to the specified device without copying storage.

train([mode])

Sets the module in training mode.

train_epoch(n_updates, input_df, df[, pbar])

Run regular epoch.

train_megabatch_epoch(n_updates, df)

Run epoch doing 'megabatch' updates, preprocessing data in large chunks.

type(dst_type)

Casts all parameters and buffers to dst_type.

xpu([device])

Moves all model parameters and buffers to the XPU.

zero_grad([set_to_none])

Sets gradients of all model parameters to zero.

__call__

build_input_tensor

compute_loss

compute_targets

create_binary_col_max

create_categorical_col_max

create_numerical_col_max

do_backward

get_anomaly_score_with_losses

get_results

get_scaler

get_variable_importance

return_feature_names

scale_losses

add_module(name, module)[source]

Adds a child module to the current module.

The module can be accessed as an attribute using the given name.

Args:
name (str): name of the child module. The child module can be

accessed from this module using the given name

module (Module): child module to be added to the module.

apply(fn)[source]

Applies fn recursively to every submodule (as returned by .children()) as well as self. Typical use includes initializing the parameters of a model (see also nn-init-doc).

Args:

fn (Module -> None): function to be applied to each submodule

Returns:

Module: self

Example:

Copy
Copied!
            

>>> @torch.no_grad() >>> def init_weights(m): >>> print(m) >>> if type(m) == nn.Linear: >>> m.weight.fill_(1.0) >>> print(m.weight) >>> net = nn.Sequential(nn.Linear(2, 2), nn.Linear(2, 2)) >>> net.apply(init_weights) Linear(in_features=2, out_features=2, bias=True) Parameter containing: tensor([[1., 1.], [1., 1.]], requires_grad=True) Linear(in_features=2, out_features=2, bias=True) Parameter containing: tensor([[1., 1.], [1., 1.]], requires_grad=True) Sequential( (0): Linear(in_features=2, out_features=2, bias=True) (1): Linear(in_features=2, out_features=2, bias=True) )

bfloat16()[source]

Casts all floating point parameters and buffers to bfloat16 datatype.

Note

This method modifies the module in-place.

Returns:

Module: self

buffers(recurse=True)[source]

Returns an iterator over module buffers.

Args:
recurse (bool): if True, then yields buffers of this module

and all submodules. Otherwise, yields only buffers that are direct members of this module.

Yields:

torch.Tensor: module buffer

Example:

Copy
Copied!
            

>>> # xdoctest: +SKIP("undefined vars") >>> for buf in model.buffers(): >>> print(type(buf), buf.size()) <class 'torch.Tensor'> (20L,) <class 'torch.Tensor'> (20L, 1L, 5L, 5L)

children()[source]

Returns an iterator over immediate children modules.

Yields:

Module: a child module

compute_baseline_performance(in_, out_)[source]
Baseline performance is computed by generating a strong

prediction for the identity function (predicting input==output) with a swapped (noisy) input, and computing the loss against the unaltered original data.

This should be roughly the loss we expect when the encoder degenerates

into the identity function solution.

Returns net loss on baseline performance computation

(sum of all losses)

compute_loss_from_targets(num, bin, cat, num_target, bin_target, cat_target, should_log=True, _id=False)[source]

Computes the loss from targets.

Parameters
numtorch.Tensor

numerical data tensor

bintorch.Tensor

binary data tensor

catList[torch.Tensor]

list of categorical data tensors

num_targettorch.Tensor

target numerical data tensor

bin_targettorch.Tensor

target binary data tensor

cat_targetList[torch.Tensor]

list of target categorical data tensors

should_logbool, optional

whether to log the loss in self.logger, by default True

_idbool, optional

whether the current step is an id validation step (for logging), by default False

Returns
Tuple[Union[float, List[float]]]

A tuple containing the mean mse/bce losses, list of mean cce losses, and mean net loss

cpu()[source]

Moves all model parameters and buffers to the CPU.

Note

This method modifies the module in-place.

Returns:

Module: self

cuda(device=None)[source]

Moves all model parameters and buffers to the GPU.

This also makes associated parameters and buffers different objects. So it should be called before constructing optimizer if the module will live on GPU while being optimized.

Note

This method modifies the module in-place.

Args:
device (int, optional): if specified, all parameters will be

copied to that device

Returns:

Module: self

decode_outputs_to_df(num, bin, cat)[source]

Converts the model outputs of the numerical, binary, and categorical features back into a pandas dataframe.

df_predict(df)[source]

Runs end-to-end model. Interprets output and creates a dataframe. Outputs dataframe with same shape as input

containing model predictions.


double()[source]

Casts all floating point parameters and buffers to double datatype.

Note

This method modifies the module in-place.

Returns:

Module: self

encode_input(df)[source]

Handles raw df inputs. Passes categories through embedding layers.

eval()[source]

Sets the module in evaluation mode.

This has any effect only on certain modules. See documentations of particular modules for details of their behaviors in training/evaluation mode, if they are affected, e.g. Dropout, BatchNorm, etc.

This is equivalent with self.train(False).

See locally-disable-grad-doc for a comparison between eval() and several similar mechanisms that may be confused with it.

Returns:

Module: self

extra_repr()[source]

Set the extra representation of the module

To print customized extra information, you should re-implement this method in your own modules. Both single-line and multi-line strings are acceptable.

fit(train_data, epochs=1, val_data=None, run_validation=False, use_val_for_loss_stats=False, rank=None, world_size=None)[source]

Does training in the specified mode (indicated by self.distrivuted_training).

Parameters
train_datapandas.DataFrame (centralized) or torch.utils.data.DataLoader (distributed)

Data for training.

epochsint, optional

Number of epochs to run training, by default 1.

val_datapandas.DataFrame (centralized) or torch.utils.data.DataLoader (distributed), optional

Data for validation and computing loss stats, by default None.

run_validationbool, optional

Whether to collect validation loss for each epoch during training, by default False.

use_val_for_loss_statsbool, optional

whether to use the validation set for loss statistics collection (for z score calculation), by default False.

rankint, optional

The rank of the current process, by default None. Required for distributed training.

world_sizeint, optional

The total number of processes, by default None. Required for distributed training.

Raises
TypeError

If train_data is not a pandas dataframe in centralized training mode.

ValueError

If rank and world_size not provided in distributed training mode.

TypeError

If train_data is not a pandas dataframe or a torch.utils.data.DataLoader or a torch.utils.data.Dataset in distributed training mode.

float()[source]

Casts all floating point parameters and buffers to float datatype.

Note

This method modifies the module in-place.

Returns:

Module: self

forward(*input)[source]

Defines the computation performed at every call.

Should be overridden by all subclasses.

get_anomaly_score(df)[source]

Returns a per-row loss of the input dataframe. Does not corrupt inputs.

get_anomaly_score_losses(df)[source]

Run the input dataframe df through the autoencoder to get the recovery losses by feature type (numerical/boolean/categorical).

get_buffer(target)[source]

Returns the buffer given by target if it exists, otherwise throws an error.

See the docstring for get_submodule for a more detailed explanation of this method’s functionality as well as how to correctly specify target.

Args:
target: The fully-qualified string name of the buffer

to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns:

torch.Tensor: The buffer referenced by target

Raises:
AttributeError: If the target string references an invalid

path or resolves to something that is not a buffer

get_deep_stack_features(df)[source]

records and outputs all internal representations of input df as row-wise vectors. Output is 2-d array with len() == len(df)

get_extra_state()[source]

Returns any extra state to include in the module’s state_dict. Implement this and a corresponding set_extra_state() for your module if you need to store extra state. This function is called when building the module’s state_dict().

Note that extra state should be pickleable to ensure working serialization of the state_dict. We only provide provide backwards compatibility guarantees for serializing Tensors; other objects may break backwards compatibility if their serialized pickled form changes.

Returns:

object: Any extra state to store in the module’s state_dict

get_parameter(target)[source]

Returns the parameter given by target if it exists, otherwise throws an error.

See the docstring for get_submodule for a more detailed explanation of this method’s functionality as well as how to correctly specify target.

Args:
target: The fully-qualified string name of the Parameter

to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns:

torch.nn.Parameter: The Parameter referenced by target

Raises:
AttributeError: If the target string references an invalid

path or resolves to something that is not an nn.Parameter

get_representation(df, layer=0)[source]
Computes latent feature vector from hidden layer

given input dataframe.

argument layer (int) specifies which layer to get. by default (layer=0), returns the “encoding” layer.

layer < 0 counts layers back from encoding layer. layer > 0 counts layers forward from encoding layer.

get_results_from_dataset(dataset, preloaded_df, return_abs=False)[source]

Returns a pandas dataframe of inference results and losses for a given dataset. Note. this function requires the whole inference set to be in loaded into memory as a pandas df

Parameters
datasettorch.utils.data.Dataset

dataset for inference

preloaded_dfpd.DataFrame

a pandas dataframe that contains the original data

return_absbool, optional

whether the absolute value of the loss scalers should be returned, by default False

Returns
pd.DataFrame

inference result with losses of each feature

get_submodule(target)[source]

Returns the submodule given by target if it exists, otherwise throws an error.

For example, let’s say you have an nn.Module A that looks like this:

Copy
Copied!
            

A( (net_b): Module( (net_c): Module( (conv): Conv2d(16, 33, kernel_size=(3, 3), stride=(2, 2)) ) (linear): Linear(in_features=100, out_features=200, bias=True) ) )

(The diagram shows an nn.Module A. A has a nested submodule net_b, which itself has two submodules net_c and linear. net_c then has a submodule conv.)

To check whether or not we have the linear submodule, we would call get_submodule("net_b.linear"). To check whether we have the conv submodule, we would call get_submodule("net_b.net_c.conv").

The runtime of get_submodule is bounded by the degree of module nesting in target. A query against named_modules achieves the same result, but it is O(N) in the number of transitive modules. So, for a simple check to see if some submodule exists, get_submodule should always be used.

Args:
target: The fully-qualified string name of the submodule

to look for. (See above example for how to specify a fully-qualified string.)

Returns:

torch.nn.Module: The submodule referenced by target

Raises:
AttributeError: If the target string references an invalid

path or resolves to something that is not an nn.Module

half()[source]

Casts all floating point parameters and buffers to half datatype.

Note

This method modifies the module in-place.

Returns:

Module: self

ipu(device=None)[source]

Moves all model parameters and buffers to the IPU.

This also makes associated parameters and buffers different objects. So it should be called before constructing optimizer if the module will live on IPU while being optimized.

Note

This method modifies the module in-place.

Arguments:
device (int, optional): if specified, all parameters will be

copied to that device

Returns:

Module: self

load_state_dict(state_dict, strict=True)[source]

Copies parameters and buffers from state_dict into this module and its descendants. If strict is True, then the keys of state_dict must exactly match the keys returned by this module’s state_dict() function.

Args:
state_dict (dict): a dict containing parameters and

persistent buffers.

strict (bool, optional): whether to strictly enforce that the keys

in state_dict match the keys returned by this module’s state_dict() function. Default: True

Returns:
NamedTuple with missing_keys and unexpected_keys fields:
  • missing_keys is a list of str containing the missing keys

  • unexpected_keys is a list of str containing the unexpected keys

Note:

If a parameter or buffer is registered as None and its corresponding key exists in state_dict, load_state_dict() will raise a RuntimeError.

modules()[source]

Returns an iterator over all modules in the network.

Yields:

Module: a module in the network

Note:

Duplicate modules are returned only once. In the following example, l will be returned only once.

Example:

Copy
Copied!
            

>>> l = nn.Linear(2, 2) >>> net = nn.Sequential(l, l) >>> for idx, m in enumerate(net.modules()): ... print(idx, '->', m) 0 -> Sequential( (0): Linear(in_features=2, out_features=2, bias=True) (1): Linear(in_features=2, out_features=2, bias=True) ) 1 -> Linear(in_features=2, out_features=2, bias=True)

named_buffers(prefix='', recurse=True)[source]

Returns an iterator over module buffers, yielding both the name of the buffer as well as the buffer itself.

Args:

prefix (str): prefix to prepend to all buffer names. recurse (bool): if True, then yields buffers of this module

and all submodules. Otherwise, yields only buffers that are direct members of this module.


Yields:

(str, torch.Tensor): Tuple containing the name and buffer

Example:

Copy
Copied!
            

>>> # xdoctest: +SKIP("undefined vars") >>> for name, buf in self.named_buffers(): >>> if name in ['running_var']: >>> print(buf.size())

named_children()[source]

Returns an iterator over immediate children modules, yielding both the name of the module as well as the module itself.

Yields:

(str, Module): Tuple containing a name and child module

Example:

Copy
Copied!
            

>>> # xdoctest: +SKIP("undefined vars") >>> for name, module in model.named_children(): >>> if name in ['conv4', 'conv5']: >>> print(module)

named_modules(memo=None, prefix='', remove_duplicate=True)[source]

Returns an iterator over all modules in the network, yielding both the name of the module as well as the module itself.

Args:

memo: a memo to store the set of modules already added to the result prefix: a prefix that will be added to the name of the module remove_duplicate: whether to remove the duplicated module instances in the result

or not


Yields:

(str, Module): Tuple of name and module

Note:

Duplicate modules are returned only once. In the following example, l will be returned only once.

Example:

Copy
Copied!
            

>>> l = nn.Linear(2, 2) >>> net = nn.Sequential(l, l) >>> for idx, m in enumerate(net.named_modules()): ... print(idx, '->', m) 0 -> ('', Sequential( (0): Linear(in_features=2, out_features=2, bias=True) (1): Linear(in_features=2, out_features=2, bias=True) )) 1 -> ('0', Linear(in_features=2, out_features=2, bias=True))

named_parameters(prefix='', recurse=True)[source]

Returns an iterator over module parameters, yielding both the name of the parameter as well as the parameter itself.

Args:

prefix (str): prefix to prepend to all parameter names. recurse (bool): if True, then yields parameters of this module

and all submodules. Otherwise, yields only parameters that are direct members of this module.


Yields:

(str, Parameter): Tuple containing the name and parameter

Example:

Copy
Copied!
            

>>> # xdoctest: +SKIP("undefined vars") >>> for name, param in self.named_parameters(): >>> if name in ['bias']: >>> print(param.size())

parameters(recurse=True)[source]

Returns an iterator over module parameters.

This is typically passed to an optimizer.

Args:
recurse (bool): if True, then yields parameters of this module

and all submodules. Otherwise, yields only parameters that are direct members of this module.

Yields:

Parameter: module parameter

Example:

Copy
Copied!
            

>>> # xdoctest: +SKIP("undefined vars") >>> for param in model.parameters(): >>> print(type(param), param.size()) <class 'torch.Tensor'> (20L,) <class 'torch.Tensor'> (20L, 1L, 5L, 5L)

prepare_df(df)[source]

Does data preparation on copy of input dataframe.

Parameters
dfpandas.DataFrame

The pandas dataframe to process

Returns
pandas.DataFrame

A processed copy of df.

preprocess_data(df, shuffle_rows_in_batch, include_original_input_tensor, include_swapped_input_by_feature_type)[source]

Preprocesses a pandas dataframe df for input into the autoencoder model.

Parameters
dfpandas.DataFrame

The input dataframe to preprocess.

shuffle_rows_in_batchbool

Whether to shuffle the rows of the dataframe before processing.

include_original_input_tensorbool

Whether to process the df into an input tensor without swapping and include it in the returned data dict. Note. Training required only the swapped input tensor while validation can use both.

include_swapped_input_by_feature_typebool

Whether to process the swapped df into num/bin/cat feature tensors and include them in the returned data dict. This is useful for baseline performance evaluation for validation.

Returns
Dict[str, Union[int, torch.Tensor]]

A dict containing the preprocessed input data and targets by feature type.

preprocess_train_data(df, shuffle_rows_in_batch=True)[source]

Wrapper function round self.preprocess_data feeding in the args suitable for a training set.

preprocess_validation_data(df, shuffle_rows_in_batch=False)[source]

Wrapper function round self.preprocess_data feeding in the args suitable for a validation set.

register_backward_hook(hook)[source]

Registers a backward hook on the module.

This function is deprecated in favor of register_full_backward_hook() and the behavior of this function will change in future versions.

Returns:
torch.utils.hooks.RemovableHandle:

a handle that can be used to remove the added hook by calling handle.remove()

register_buffer(name, tensor, persistent=True)[source]

Adds a buffer to the module.

This is typically used to register a buffer that should not to be considered a model parameter. For example, BatchNorm’s running_mean is not a parameter, but is part of the module’s state. Buffers, by default, are persistent and will be saved alongside parameters. This behavior can be changed by setting persistent to False. The only difference between a persistent buffer and a non-persistent buffer is that the latter will not be a part of this module’s state_dict.

Buffers can be accessed as attributes using given names.

Args:
name (str): name of the buffer. The buffer can be accessed

from this module using the given name

tensor (Tensor or None): buffer to be registered. If None, then operations

that run on buffers, such as cuda, are ignored. If None, the buffer is not included in the module’s state_dict.

persistent (bool): whether the buffer is part of this module’s

state_dict.

Example:

Copy
Copied!
            

>>> # xdoctest: +SKIP("undefined vars") >>> self.register_buffer('running_mean', torch.zeros(num_features))

register_forward_hook(hook)[source]

Registers a forward hook on the module.

The hook will be called every time after forward() has computed an output. It should have the following signature:

Copy
Copied!
            

hook(module, input, output) -> None or modified output

The input contains only the positional arguments given to the module. Keyword arguments won’t be passed to the hooks and only to the forward. The hook can modify the output. It can modify the input inplace but it will not have effect on forward since this is called after forward() is called.

Returns:
torch.utils.hooks.RemovableHandle:

a handle that can be used to remove the added hook by calling handle.remove()

register_forward_pre_hook(hook)[source]

Registers a forward pre-hook on the module.

The hook will be called every time before forward() is invoked. It should have the following signature:

Copy
Copied!
            

hook(module, input) -> None or modified input

The input contains only the positional arguments given to the module. Keyword arguments won’t be passed to the hooks and only to the forward. The hook can modify the input. User can either return a tuple or a single modified value in the hook. We will wrap the value into a tuple if a single value is returned(unless that value is already a tuple).

Returns:
torch.utils.hooks.RemovableHandle:

a handle that can be used to remove the added hook by calling handle.remove()

register_full_backward_hook(hook)[source]

Registers a backward hook on the module.

The hook will be called every time the gradients with respect to a module are computed, i.e. the hook will execute if and only if the gradients with respect to module outputs are computed. The hook should have the following signature:

Copy
Copied!
            

hook(module, grad_input, grad_output) -> tuple(Tensor) or None

The grad_input and grad_output are tuples that contain the gradients with respect to the inputs and outputs respectively. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the input that will be used in place of grad_input in subsequent computations. grad_input will only correspond to the inputs given as positional arguments and all kwarg arguments are ignored. Entries in grad_input and grad_output will be None for all non-Tensor arguments.

For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module’s forward function.

Warning

Modifying inputs or outputs inplace is not allowed when using backward hooks and will raise an error.

Returns:
torch.utils.hooks.RemovableHandle:

a handle that can be used to remove the added hook by calling handle.remove()

register_load_state_dict_post_hook(hook)[source]

Registers a post hook to be run after module’s load_state_dict is called.

It should have the following signature::

hook(module, incompatible_keys) -> None

The module argument is the current module that this hook is registered on, and the incompatible_keys argument is a NamedTuple consisting of attributes missing_keys and unexpected_keys. missing_keys is a list of str containing the missing keys and unexpected_keys is a list of str containing the unexpected keys.

The given incompatible_keys can be modified inplace if needed.

Note that the checks performed when calling load_state_dict() with strict=True are affected by modifications the hook makes to missing_keys or unexpected_keys, as expected. Additions to either set of keys will result in an error being thrown when strict=True, and clearning out both missing and unexpected keys will avoid an error.

Returns:
torch.utils.hooks.RemovableHandle:

a handle that can be used to remove the added hook by calling handle.remove()

register_module(name, module)[source]

Alias for add_module().

register_parameter(name, param)[source]

Adds a parameter to the module.

The parameter can be accessed as an attribute using given name.

Args:
name (str): name of the parameter. The parameter can be accessed

from this module using the given name

param (Parameter or None): parameter to be added to the module. If

None, then operations that run on parameters, such as cuda, are ignored. If None, the parameter is not included in the module’s state_dict.

requires_grad_(requires_grad=True)[source]

Change if autograd should record operations on parameters in this module.

This method sets the parameters’ requires_grad attributes in-place.

This method is helpful for freezing part of the module for finetuning or training parts of a model individually (e.g., GAN training).

See locally-disable-grad-doc for a comparison between requires_grad_() and several similar mechanisms that may be confused with it.

Args:
requires_grad (bool): whether autograd should record operations on

parameters in this module. Default: True.

Returns:

Module: self

set_extra_state(state)[source]

This function is called from load_state_dict() to handle any extra state found within the state_dict. Implement this function and a corresponding get_extra_state() for your module if you need to store extra state within its state_dict.

Args:

state (dict): Extra state from the state_dict

share_memory()[source]

See torch.Tensor.share_memory_()

state_dict(*args, destination=None, prefix='', keep_vars=False)[source]

Returns a dictionary containing references to the whole state of the module.

Both parameters and persistent buffers (e.g. running averages) are included. Keys are corresponding parameter and buffer names. Parameters and buffers set to None are not included.

Note

The returned object is a shallow copy. It contains references to the module’s parameters and buffers.

Warning

Currently state_dict() also accepts positional arguments for destination, prefix and keep_vars in order. However, this is being deprecated and keyword arguments will be enforced in future releases.

Warning

Please avoid the use of argument destination as it is not designed for end-users.

Args:
destination (dict, optional): If provided, the state of module will

be updated into the dict and the same object is returned. Otherwise, an OrderedDict will be created and returned. Default: None.

prefix (str, optional): a prefix added to parameter and buffer

names to compose the keys in state_dict. Default: ''.

keep_vars (bool, optional): by default the Tensor s

returned in the state dict are detached from autograd. If it’s set to True, detaching will not be performed. Default: False.

Returns:
dict:

a dictionary containing a whole state of the module

Example:

Copy
Copied!
            

>>> # xdoctest: +SKIP("undefined vars") >>> module.state_dict().keys() ['bias', 'weight']

to(*args, **kwargs)[source]

Moves and/or casts the parameters and buffers.

This can be called as

to(device=None, dtype=None, non_blocking=False)[source]

to(dtype, non_blocking=False)[source]

to(tensor, non_blocking=False)[source]

to(memory_format=torch.channels_last)[source]

Its signature is similar to torch.Tensor.to(), but only accepts floating point or complex dtypes. In addition, this method will only cast the floating point or complex parameters and buffers to dtype (if given). The integral parameters and buffers will be moved device, if that is given, but with dtypes unchanged. When non_blocking is set, it tries to convert/move asynchronously with respect to the host if possible, e.g., moving CPU Tensors with pinned memory to CUDA devices.

See below for examples.

Note

This method modifies the module in-place.

Args:
device (torch.device): the desired device of the parameters

and buffers in this module

dtype (torch.dtype): the desired floating point or complex dtype of

the parameters and buffers in this module

tensor (torch.Tensor): Tensor whose dtype and device are the desired

dtype and device for all parameters and buffers in this module

memory_format (torch.memory_format): the desired memory

format for 4D parameters and buffers in this module (keyword only argument)

Returns:

Module: self

Examples:

Copy
Copied!
            

>>> # xdoctest: +IGNORE_WANT("non-deterministic") >>> linear = nn.Linear(2, 2) >>> linear.weight Parameter containing: tensor([[ 0.1913, -0.3420], [-0.5113, -0.2325]]) >>> linear.to(torch.double) Linear(in_features=2, out_features=2, bias=True) >>> linear.weight Parameter containing: tensor([[ 0.1913, -0.3420], [-0.5113, -0.2325]], dtype=torch.float64) >>> # xdoctest: +REQUIRES(env:TORCH_DOCTEST_CUDA1) >>> gpu1 = torch.device("cuda:1") >>> linear.to(gpu1, dtype=torch.half, non_blocking=True) Linear(in_features=2, out_features=2, bias=True) >>> linear.weight Parameter containing: tensor([[ 0.1914, -0.3420], [-0.5112, -0.2324]], dtype=torch.float16, device='cuda:1') >>> cpu = torch.device("cpu") >>> linear.to(cpu) Linear(in_features=2, out_features=2, bias=True) >>> linear.weight Parameter containing: tensor([[ 0.1914, -0.3420], [-0.5112, -0.2324]], dtype=torch.float16) >>> linear = nn.Linear(2, 2, bias=None).to(torch.cdouble) >>> linear.weight Parameter containing: tensor([[ 0.3741+0.j, 0.2382+0.j], [ 0.5593+0.j, -0.4443+0.j]], dtype=torch.complex128) >>> linear(torch.ones(3, 2, dtype=torch.cdouble)) tensor([[0.6122+0.j, 0.1150+0.j], [0.6122+0.j, 0.1150+0.j], [0.6122+0.j, 0.1150+0.j]], dtype=torch.complex128)

to_empty(*, device)[source]

Moves the parameters and buffers to the specified device without copying storage.

Args:
device (torch.device): The desired device of the parameters

and buffers in this module.

Returns:

Module: self

train(mode=True)[source]

Sets the module in training mode.

This has any effect only on certain modules. See documentations of particular modules for details of their behaviors in training/evaluation mode, if they are affected, e.g. Dropout, BatchNorm, etc.

Args:
mode (bool): whether to set training mode (True) or evaluation

mode (False). Default: True.

Returns:

Module: self

train_epoch(n_updates, input_df, df, pbar=None)[source]

Run regular epoch.

train_megabatch_epoch(n_updates, df)[source]

Run epoch doing ‘megabatch’ updates, preprocessing data in large chunks.

type(dst_type)[source]

Casts all parameters and buffers to dst_type.

Note

This method modifies the module in-place.

Args:

dst_type (type or string): the desired type

Returns:

Module: self

xpu(device=None)[source]

Moves all model parameters and buffers to the XPU.

This also makes associated parameters and buffers different objects. So it should be called before constructing optimizer if the module will live on XPU while being optimized.

Note

This method modifies the module in-place.

Arguments:
device (int, optional): if specified, all parameters will be

copied to that device

Returns:

Module: self

zero_grad(set_to_none=False)[source]

Sets gradients of all model parameters to zero. See similar function under torch.optim.Optimizer for more context.

Args:
set_to_none (bool): instead of setting to zero, set the grads to None.

See torch.optim.Optimizer.zero_grad() for details.

© Copyright 2023, NVIDIA. Last updated on Apr 11, 2023.