ParallelEnv

class torchrl.envs.ParallelEnv(*args, _inplace_update=False, _batch_locked=True, **kwargs)

Creates one environment per process.

TensorDicts are passed via shared memory or memory map.

Batched environments allow the user to query an arbitrary method / attribute of the environment running remotely.

Those queries will return a list of length equal to the number of workers containing the values resulting from those queries.

>>> env = ParallelEnv(3, my_env_fun)
>>> custom_attribute_list = env.custom_attribute
>>> custom_method_list = env.custom_method(*args)

Parameters:

• num_workers – number of workers (i.e. env instances) to be deployed simultaneously;

• create_env_fn (callable or list of callables) – function (or list of functions) to be used for the environment creation. If a single task is used, a callable should be used and not a list of identical callables: if a list of callable is provided, the environment will be executed as if multiple, diverse tasks were needed, which comes with a slight compute overhead;

• create_env_kwargs (dict or list of dicts, optional) – kwargs to be used with the environments being created;

• pin_memory (bool) – if True and device is “cpu”, calls pin_memory on the tensordicts when created.

• share_individual_td (bool, optional) – if True, a different tensordict is created for every process/worker and a lazy stack is returned. default = None (False if single task);

• shared_memory (bool) – whether or not the returned tensordict will be placed in shared memory;

• memmap (bool) – whether or not the returned tensordict will be placed in memory map.

• policy_proof (callable, optional) – if provided, it’ll be used to get the list of tensors to return through the step() and reset() methods, such as "hidden" etc.

• device (str, int, torch.device) – for consistency, this argument is kept. However this argument should not be passed, as the device will be inferred from the environments. It is assumed that all environments will run on the same device as a common shared tensordict will be used to pass data from process to process. The device can be changed after instantiation using env.to(device).

• allow_step_when_done (bool, optional) – if True, batched environments can execute steps after a done state is encountered. Defaults to False.

property action_key

The action key of an environment.

By default, non-nested keys are stored in the ‘action’ key.

If the action is in a nested tensordict, this property will return its location.

property action_spec: TensorSpec

The action leaf spec.

This property will always return the leaf spec of the action attribute, which can be accessed in a typical rollout via

>>> fake_td = env.fake_tensordict()  # a typical tensordict
>>> action = fake_td[env.action_key]

This property is mutable.

add_module(name: str, module: Module | None) → None

Adds a child module to the current module.

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

Parameters:

• 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: Callable[[Module], None]) → T

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 torch.nn.init).

Parameters:

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

Returns:

self

Return type:

Module

Example:

>>> @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)
)

property batch_locked: bool

Whether the environnement can be used with a batch size different from the one it was initialized with or not.

If True, the env needs to be used with a tensordict having the same batch size as the env. batch_locked is an immutable property.

bfloat16() → T

Casts all floating point parameters and buffers to bfloat16 datatype.

Note

This method modifies the module in-place.

Returns:

self

Return type:

Module

buffers(recurse: bool = True) → Iterator[Tensor]

Returns an iterator over module buffers.

Parameters:

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:

>>> # 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() → Iterator[Module]

Returns an iterator over immediate children modules.

Yields:

Module – a child module

compile(*args, **kwargs)

Compile this Module’s forward using torch.compile().

This Module’s __call__ method is compiled and all arguments are passed as-is to torch.compile().

See torch.compile() for details on the arguments for this function.

cpu() → T

Moves all model parameters and buffers to the CPU.

Note

This method modifies the module in-place.

Returns:

self

Return type:

Module

cuda(device: int | device | None = None) → T

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.

Parameters:

device (int, optional) – if specified, all parameters will be copied to that device

Returns:

self

Return type:

Module

property done_key

The done key of an environment.

By default, non-nested keys are stored in the 'done' entry.

If the done is in a nested tensordict, this property will return its location.

property done_spec: TensorSpec

The done leaf spec.

This property will always return the leaf spec of the done attribute, which can be accessed in a typical rollout via

>>> fake_td = env.fake_tensordict()  # a typical tensordict
>>> done = fake_td[("next", *env.done_key)]

This property is mutable.

double() → T

Casts all floating point parameters and buffers to double datatype.

Note

This method modifies the module in-place.

Returns:

self

Return type:

Module

eval() → T

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 disabling gradient computation for a comparison between .eval() and several similar mechanisms that may be confused with it.

Returns:

self

Return type:

Module

extra_repr() → str

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.

fake_tensordict() → TensorDictBase

Returns a fake tensordict with key-value pairs that match in shape, device and dtype what can be expected during an environment rollout.

float() → T

Casts all floating point parameters and buffers to float datatype.

Note

This method modifies the module in-place.

Returns:

self

Return type:

Module

forward(tensordict: TensorDictBase) → TensorDictBase

Defines 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.

get_buffer(target: str) → Tensor

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.

Parameters:

target – The fully-qualified string name of the buffer to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns:

The buffer referenced by target

Return type:

torch.Tensor

Raises:

AttributeError – If the target string references an invalid path or resolves to something that is not a buffer

get_extra_state() → Any

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 picklable 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:

Any extra state to store in the module’s state_dict

Return type:

object

get_parameter(target: str) → Parameter

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.

Parameters:

target – The fully-qualified string name of the Parameter to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns:

The Parameter referenced by target

Return type:

torch.nn.Parameter

Raises:

AttributeError – If the target string references an invalid path or resolves to something that is not an nn.Parameter

get_submodule(target: str) → Module

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:

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.

Parameters:

target – The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.)

Returns:

The submodule referenced by target

Return type:

torch.nn.Module

Raises:

AttributeError – If the target string references an invalid path or resolves to something that is not an nn.Module

half() → T

Casts all floating point parameters and buffers to half datatype.

Note

This method modifies the module in-place.

Returns:

self

Return type:

Module

ipu(device: int | device | None = None) → T

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.

Parameters:

device (int, optional) – if specified, all parameters will be copied to that device

Returns:

self

Return type:

Module

load_state_dict(*args, **kwargs)

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.

Parameters:

• 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:

• missing_keys is a list of str containing the missing keys

• unexpected_keys is a list of str containing the unexpected keys

Return type:

NamedTuple with missing_keys and unexpected_keys fields

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() → Iterator[Module]

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:

>>> 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: str = '', recurse: bool = True, remove_duplicate: bool = True) → Iterator[Tuple[str, Tensor]]

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

Parameters:

• prefix (str) – prefix to prepend to all buffer names.

• recurse (bool, optional) – if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module. Defaults to True.

• remove_duplicate (bool, optional) – whether to remove the duplicated buffers in the result. Defaults to True.

Yields:

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

Example:

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

named_children() → Iterator[Tuple[str, Module]]

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:

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

named_modules(memo: Set[Module] | None = None, prefix: str = '', remove_duplicate: bool = True)

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

Parameters:

• 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:

>>> 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: str = '', recurse: bool = True, remove_duplicate: bool = True) → Iterator[Tuple[str, Parameter]]

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

Parameters:

• 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.

• remove_duplicate (bool, optional) – whether to remove the duplicated parameters in the result. Defaults to True.

Yields:

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

Example:

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

parameters(recurse: bool = True) → Iterator[Parameter]

Returns an iterator over module parameters.

This is typically passed to an optimizer.

Parameters:

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:

>>> # 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)

rand_action(tensordict: TensorDictBase | None = None)

Performs a random action given the action_spec attribute.

Parameters:

tensordict (TensorDictBase, optional) – tensordict where the resulting action should be written.

Returns:

a tensordict object with the “action” entry updated with a random sample from the action-spec.

rand_step(tensordict: TensorDictBase | None = None) → TensorDictBase

Performs a random step in the environment given the action_spec attribute.

Parameters:

tensordict (TensorDictBase, optional) – tensordict where the resulting info should be written.

Returns:

a tensordict object with the new observation after a random step in the environment. The action will be stored with the “action” key.

register_backward_hook(hook: Callable[[Module, Tuple[Tensor, ...] | Tensor, Tuple[Tensor, ...] | Tensor], None | Tuple[Tensor, ...] | Tensor]) → RemovableHandle

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:

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

Return type:

torch.utils.hooks.RemovableHandle

register_buffer(name: str, tensor: Tensor | None, persistent: bool = True) → None

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.

Parameters:

• 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:

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

register_forward_hook(hook: Callable[[T, Tuple[Any, ...], Any], Any | None] | Callable[[T, Tuple[Any, ...], Dict[str, Any], Any], Any | None], *, prepend: bool = False, with_kwargs: bool = False) → RemovableHandle

Registers a forward hook on the module.

The hook will be called every time after forward() has computed an output.

If with_kwargs is False or not specified, 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. The hook should have the following signature:

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

If with_kwargs is True, the forward hook will be passed the kwargs given to the forward function and be expected to return the output possibly modified. The hook should have the following signature:

hook(module, args, kwargs, output) -> None or modified output

Parameters:

• hook (Callable) – The user defined hook to be registered.

• prepend (bool) – If True, the provided hook will be fired before all existing forward hooks on this torch.nn.modules.Module. Otherwise, the provided hook will be fired after all existing forward hooks on this torch.nn.modules.Module. Note that global forward hooks registered with register_module_forward_hook() will fire before all hooks registered by this method. Default: False

• with_kwargs (bool) – If True, the hook will be passed the kwargs given to the forward function. Default: False

Returns:

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

Return type:

torch.utils.hooks.RemovableHandle

register_forward_pre_hook(hook: Callable[[T, Tuple[Any, ...]], Any | None] | Callable[[T, Tuple[Any, ...], Dict[str, Any]], Tuple[Any, Dict[str, Any]] | None], *, prepend: bool = False, with_kwargs: bool = False) → RemovableHandle

Registers a forward pre-hook on the module.

The hook will be called every time before forward() is invoked.

If with_kwargs is false or not specified, 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). The hook should have the following signature:

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

If with_kwargs is true, the forward pre-hook will be passed the kwargs given to the forward function. And if the hook modifies the input, both the args and kwargs should be returned. The hook should have the following signature:

hook(module, args, kwargs) -> None or a tuple of modified input and kwargs

Parameters:

• hook (Callable) – The user defined hook to be registered.

• prepend (bool) – If true, the provided hook will be fired before all existing forward_pre hooks on this torch.nn.modules.Module. Otherwise, the provided hook will be fired after all existing forward_pre hooks on this torch.nn.modules.Module. Note that global forward_pre hooks registered with register_module_forward_pre_hook() will fire before all hooks registered by this method. Default: False

• with_kwargs (bool) – If true, the hook will be passed the kwargs given to the forward function. Default: False

Returns:

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

Return type:

torch.utils.hooks.RemovableHandle

register_full_backward_hook(hook: Callable[[Module, Tuple[Tensor, ...] | Tensor, Tuple[Tensor, ...] | Tensor], None | Tuple[Tensor, ...] | Tensor], prepend: bool = False) → RemovableHandle

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:

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.

Parameters:

• hook (Callable) – The user-defined hook to be registered.

• prepend (bool) – If true, the provided hook will be fired before all existing backward hooks on this torch.nn.modules.Module. Otherwise, the provided hook will be fired after all existing backward hooks on this torch.nn.modules.Module. Note that global backward hooks registered with register_module_full_backward_hook() will fire before all hooks registered by this method.

Returns:

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

Return type:

torch.utils.hooks.RemovableHandle

register_full_backward_pre_hook(hook: Callable[[Module, Tuple[Tensor, ...] | Tensor], None | Tuple[Tensor, ...] | Tensor], prepend: bool = False) → RemovableHandle

Registers a backward pre-hook on the module.

The hook will be called every time the gradients for the module are computed. The hook should have the following signature:

hook(module, grad_output) -> Tensor or None

The grad_output is a tuple. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the output that will be used in place of grad_output in subsequent computations. Entries in 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 inplace is not allowed when using backward hooks and will raise an error.

Parameters:

• hook (Callable) – The user-defined hook to be registered.

• prepend (bool) – If true, the provided hook will be fired before all existing backward_pre hooks on this torch.nn.modules.Module. Otherwise, the provided hook will be fired after all existing backward_pre hooks on this torch.nn.modules.Module. Note that global backward_pre hooks registered with register_module_full_backward_pre_hook() will fire before all hooks registered by this method.

Returns:

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

Return type:

torch.utils.hooks.RemovableHandle

register_load_state_dict_post_hook(hook)

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 clearing out both missing and unexpected keys will avoid an error.

Returns:

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

Return type:

torch.utils.hooks.RemovableHandle

register_module(name: str, module: Module | None) → None

Alias for add_module().

register_parameter(name: str, param: Parameter | None) → None

Adds a parameter to the module.

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

Parameters:

• 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.

register_state_dict_pre_hook(hook)

These hooks will be called with arguments: self, prefix, and keep_vars before calling state_dict on self. The registered hooks can be used to perform pre-processing before the state_dict call is made.

requires_grad_(requires_grad: bool = True) → T

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 disabling gradient computation for a comparison between .requires_grad_() and several similar mechanisms that may be confused with it.

Parameters:

requires_grad (bool) – whether autograd should record operations on parameters in this module. Default: True.

Returns:

self

Return type:

Module

reset(tensordict: TensorDictBase | None = None, **kwargs) → TensorDictBase

Resets the environment.

As for step and _step, only the private method _reset should be overwritten by EnvBase subclasses.

Parameters:

• tensordict (TensorDictBase, optional) – tensordict to be used to contain the resulting new observation. In some cases, this input can also be used to pass argument to the reset function.

• kwargs (optional) – other arguments to be passed to the native reset function.

Returns:

a tensordict (or the input tensordict, if any), modified in place with the resulting observations.

property reward_key

The reward key of an environment.

By default, non-nested keys are stored in the 'reward' entry.

If the reward is in a nested tensordict, this property will return its location.

property reward_spec: TensorSpec

The reward leaf spec.

This property will always return the leaf spec of the reward attribute, which can be accessed in a typical rollout via

>>> fake_td = env.fake_tensordict()  # a typical tensordict
>>> reward = fake_td[("next", *env.reward_key)]

This property is mutable.

rollout(max_steps: int, policy: Callable[[TensorDictBase], TensorDictBase] | None = None, callback: Callable[[TensorDictBase, ...], TensorDictBase] | None = None, auto_reset: bool = True, auto_cast_to_device: bool = False, break_when_any_done: bool = True, return_contiguous: bool = True, tensordict: TensorDictBase | None = None) → TensorDictBase

Executes a rollout in the environment.

The function will stop as soon as one of the contained environments returns done=True.

Parameters:

• max_steps (int) – maximum number of steps to be executed. The actual number of steps can be smaller if the environment reaches a done state before max_steps have been executed.

• policy (callable, optional) – callable to be called to compute the desired action. If no policy is provided, actions will be called using env.rand_step() default = None

• callback (callable, optional) – function to be called at each iteration with the given TensorDict.

• auto_reset (bool, optional) – if True, resets automatically the environment if it is in a done state when the rollout is initiated. Default is True.

• auto_cast_to_device (bool, optional) – if True, the device of the tensordict is automatically cast to the policy device before the policy is used. Default is False.

• break_when_any_done (bool) – breaks if any of the done state is True. If False, a reset() is called on the sub-envs that are done. Default is True.

• return_contiguous (bool) – if False, a LazyStackedTensorDict will be returned. Default is True.

• tensordict (TensorDict, optional) – if auto_reset is False, an initial tensordict must be provided.

Returns:

TensorDict object containing the resulting trajectory.

The data returned will be marked with a “time” dimension name for the last dimension of the tensordict (at the env.ndim index).

Examples

>>> from torchrl.envs.libs.gym import GymEnv
>>> from torchrl.envs.transforms import TransformedEnv, StepCounter
>>> env = TransformedEnv(GymEnv("Pendulum-v1"), StepCounter(max_steps=20))
>>> rollout = env.rollout(max_steps=1000)
>>> print(rollout)
TensorDict(
    fields={
        action: Tensor(shape=torch.Size([20, 1]), device=cpu, dtype=torch.float32, is_shared=False),
        done: Tensor(shape=torch.Size([20, 1]), device=cpu, dtype=torch.bool, is_shared=False),
        next: TensorDict(
            fields={
                done: Tensor(shape=torch.Size([20, 1]), device=cpu, dtype=torch.bool, is_shared=False),
                observation: Tensor(shape=torch.Size([20, 3]), device=cpu, dtype=torch.float32, is_shared=False),
                reward: Tensor(shape=torch.Size([20, 1]), device=cpu, dtype=torch.float32, is_shared=False),
                step_count: Tensor(shape=torch.Size([20, 1]), device=cpu, dtype=torch.int64, is_shared=False),
                truncated: Tensor(shape=torch.Size([20, 1]), device=cpu, dtype=torch.bool, is_shared=False)},
            batch_size=torch.Size([20]),
            device=cpu,
            is_shared=False),
        observation: Tensor(shape=torch.Size([20, 3]), device=cpu, dtype=torch.float32, is_shared=False),
        step_count: Tensor(shape=torch.Size([20, 1]), device=cpu, dtype=torch.int64, is_shared=False),
        truncated: Tensor(shape=torch.Size([20, 1]), device=cpu, dtype=torch.bool, is_shared=False)},
    batch_size=torch.Size([20]),
    device=cpu,
    is_shared=False)
>>> print(rollout.names)
['time']
>>> # with envs that contain more dimensions
>>> from torchrl.envs import SerialEnv
>>> env = SerialEnv(3, lambda: TransformedEnv(GymEnv("Pendulum-v1"), StepCounter(max_steps=20)))
>>> rollout = env.rollout(max_steps=1000)
>>> print(rollout)
TensorDict(
    fields={
        action: Tensor(shape=torch.Size([3, 20, 1]), device=cpu, dtype=torch.float32, is_shared=False),
        done: Tensor(shape=torch.Size([3, 20, 1]), device=cpu, dtype=torch.bool, is_shared=False),
        next: TensorDict(
            fields={
                done: Tensor(shape=torch.Size([3, 20, 1]), device=cpu, dtype=torch.bool, is_shared=False),
                observation: Tensor(shape=torch.Size([3, 20, 3]), device=cpu, dtype=torch.float32, is_shared=False),
                reward: Tensor(shape=torch.Size([3, 20, 1]), device=cpu, dtype=torch.float32, is_shared=False),
                step_count: Tensor(shape=torch.Size([3, 20, 1]), device=cpu, dtype=torch.int64, is_shared=False),
                truncated: Tensor(shape=torch.Size([3, 20, 1]), device=cpu, dtype=torch.bool, is_shared=False)},
            batch_size=torch.Size([3, 20]),
            device=cpu,
            is_shared=False),
        observation: Tensor(shape=torch.Size([3, 20, 3]), device=cpu, dtype=torch.float32, is_shared=False),
        step_count: Tensor(shape=torch.Size([3, 20, 1]), device=cpu, dtype=torch.int64, is_shared=False),
        truncated: Tensor(shape=torch.Size([3, 20, 1]), device=cpu, dtype=torch.bool, is_shared=False)},
    batch_size=torch.Size([3, 20]),
    device=cpu,
    is_shared=False)
>>> print(rollout.names)
[None, 'time']

In some instances, contiguous tensordict cannot be obtained because they cannot be stacked. This can happen when the data returned at each step may have a different shape, or when different environments are executed together. In that case, return_contiguous=False will cause the returned tensordict to be a lazy stack of tensordicts:

Examples

>>> rollout = env.rollout(4, return_contiguous=False)
>>> print(rollout)

LazyStackedTensorDict(

fields={

action: Tensor(shape=torch.Size([3, 4, 1]), device=cpu, dtype=torch.float32, is_shared=False), done: Tensor(shape=torch.Size([3, 4, 1]), device=cpu, dtype=torch.bool, is_shared=False), next: LazyStackedTensorDict(

fields={

done: Tensor(shape=torch.Size([3, 4, 1]), device=cpu, dtype=torch.bool, is_shared=False), observation: Tensor(shape=torch.Size([3, 4, 3]), device=cpu, dtype=torch.float32, is_shared=False), reward: Tensor(shape=torch.Size([3, 4, 1]), device=cpu, dtype=torch.float32, is_shared=False), step_count: Tensor(shape=torch.Size([3, 4, 1]), device=cpu, dtype=torch.int64, is_shared=False), truncated: Tensor(shape=torch.Size([3, 4, 1]), device=cpu, dtype=torch.bool, is_shared=False)},

batch_size=torch.Size([3, 4]), device=cpu, is_shared=False),

observation: Tensor(shape=torch.Size([3, 4, 3]), device=cpu, dtype=torch.float32, is_shared=False), step_count: Tensor(shape=torch.Size([3, 4, 1]), device=cpu, dtype=torch.int64, is_shared=False), truncated: Tensor(shape=torch.Size([3, 4, 1]), device=cpu, dtype=torch.bool, is_shared=False)},

batch_size=torch.Size([3, 4]), device=cpu, is_shared=False) >>> print(rollout.names) [None, ‘time’]

set_extra_state(state: Any)

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.

Parameters:

state (dict) – Extra state from the state_dict

set_seed(*args, **kwargs)

Sets the seed of the environment and returns the next seed to be used (which is the input seed if a single environment is present).

Parameters:

• seed (int) – seed to be set

• static_seed (bool, optional) – if True, the seed is not incremented. Defaults to False

Returns:

i.e. the seed that should be used for another environment if created concomittently to this environment.

Return type:

integer representing the “next seed”

share_memory() → T

See torch.Tensor.share_memory_()

property specs: CompositeSpec

Returns a Composite container where all the environment are present.

This feature allows one to create an environment, retrieve all of the specs in a single data container and then erase the environment from the workspace.

state_dict(*args, **kwargs)

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.

Parameters:

• 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:

a dictionary containing a whole state of the module

Return type:

dict

Example:

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

step(tensordict: TensorDictBase) → TensorDictBase

Makes a step in the environment.

Step accepts a single argument, tensordict, which usually carries an ‘action’ key which indicates the action to be taken. Step will call an out-place private method, _step, which is the method to be re-written by EnvBase subclasses.

Parameters:

tensordict (TensorDictBase) – Tensordict containing the action to be taken.

Returns:

the input tensordict, modified in place with the resulting observations, done state and reward (+ others if needed).

to(device: device | str | int)

Moves and/or casts the parameters and buffers.

This can be called as

to(device=None, dtype=None, non_blocking=False)

to(dtype, non_blocking=False)

to(tensor, non_blocking=False)

to(memory_format=torch.channels_last)

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.

Parameters:

• 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:

self

Return type:

Module

Examples:

>>> # 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: str | device) → T

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

Parameters:

device (torch.device) – The desired device of the parameters and buffers in this module.

Returns:

self

Return type:

Module

train(mode: bool = True) → T

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.

Parameters:

mode (bool) – whether to set training mode (True) or evaluation mode (False). Default: True.

Returns:

self

Return type:

Module

type(dst_type: dtype | str) → T

Casts all parameters and buffers to dst_type.

Note

This method modifies the module in-place.

Parameters:

dst_type (type or string) – the desired type

Returns:

self

Return type:

Module

update_kwargs(kwargs: dict | List[dict]) → None

Updates the kwargs of each environment given a dictionary or a list of dictionaries.

Parameters:

kwargs (dict or list of dict) – new kwargs to use with the environments

xpu(device: int | device | None = None) → T

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.

Parameters:

device (int, optional) – if specified, all parameters will be copied to that device

Returns:

self

Return type:

Module

zero_grad(set_to_none: bool = True) → None

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

Parameters:

set_to_none (bool) – instead of setting to zero, set the grads to None. See torch.optim.Optimizer.zero_grad() for details.