RewardData¶

class torchrl.data.RewardData(input_ids: 'torch.Tensor', attention_mask: 'torch.Tensor', rewards: 'Optional[torch.Tensor]' = None, end_scores: 'Optional[torch.Tensor]' = None, *, batch_size, device=None, names=None)[source]¶

property batch_size: Size¶

Retrieves the batch size for the tensor class.

Returns:: batch size (torch.Size)

property device: device¶: Retrieves the device type of tensor class.

classmethod from_tensordict(tensordict, non_tensordict=None)¶

Tensor class wrapper to instantiate a new tensor class object.

Parameters:

tensordict (TensorDict) – Dictionary of tensor types
non_tensordict (dict) – Dictionary with non-tensor and nested tensor class objects

get(key: ~typing.Union[str, ~typing.Tuple[str, ...]], default: ~typing.Any = <tensordict.base._NoDefault object>)¶

Gets the value stored with the input key.

Parameters:

key (str, tuple of str) – key to be queried. If tuple of str it is equivalent to chained calls of getattr.
default – default value if the key is not found in the tensorclass.

Returns:

value stored with the input key

classmethod load_memmap(prefix: str | Path, device: torch.device | None = None, non_blocking: bool = False) → T¶

Loads a memory-mapped tensordict from disk.

Parameters:

prefix (str or Path to folder) – the path to the folder where the saved tensordict should be fetched.
device (torch.device or equivalent, optional) – if provided, the data will be asynchronously cast to that device. Supports “meta” device, in which case the data isn’t loaded but a set of empty “meta” tensors are created. This is useful to get a sense of the total model size and structure without actually opening any file.
non_blocking (bool, optional) – if True, synchronize won’t be called after loading tensors on device. Defaults to False.

Examples

>>> from tensordict import TensorDict
>>> td = TensorDict.fromkeys(["a", "b", "c", ("nested", "e")], 0)
>>> td.memmap("./saved_td")
>>> td_load = TensorDict.load_memmap("./saved_td")
>>> assert (td == td_load).all()

This method also allows loading nested tensordicts.

>>> nested = TensorDict.load_memmap("./saved_td/nested")
>>> assert nested["e"] == 0

A tensordict can also be loaded on “meta” device or, alternatively, as a fake tensor:

>>> import tempfile
>>> td = TensorDict({"a": torch.zeros(()), "b": {"c": torch.zeros(())}})
>>> with tempfile.TemporaryDirectory() as path:
...     td.save(path)
...     td_load = TensorDict.load_memmap(path, device="meta")
...     print("meta:", td_load)
...     from torch._subclasses import FakeTensorMode
...     with FakeTensorMode():
...         td_load = TensorDict.load_memmap(path)
...         print("fake:", td_load)
meta: TensorDict(
    fields={
        a: Tensor(shape=torch.Size([]), device=meta, dtype=torch.float32, is_shared=False),
        b: TensorDict(
            fields={
                c: Tensor(shape=torch.Size([]), device=meta, dtype=torch.float32, is_shared=False)},
            batch_size=torch.Size([]),
            device=meta,
            is_shared=False)},
    batch_size=torch.Size([]),
    device=meta,
    is_shared=False)
fake: TensorDict(
    fields={
        a: FakeTensor(shape=torch.Size([]), device=cpu, dtype=torch.float32, is_shared=False),
        b: TensorDict(
            fields={
                c: FakeTensor(shape=torch.Size([]), device=cpu, dtype=torch.float32, is_shared=False)},
            batch_size=torch.Size([]),
            device=cpu,
            is_shared=False)},
    batch_size=torch.Size([]),
    device=cpu,
    is_shared=False)

load_state_dict(state_dict: dict[str, Any], strict=True, assign=False, from_flatten=False)¶: Loads a state_dict attemptedly in-place on the destination tensorclass.

memmap(prefix: str | None = None, copy_existing: bool = False, *, num_threads: int = 0, return_early: bool = False, share_non_tensor: bool = False) → T¶

Writes all tensors onto a corresponding memory-mapped Tensor in a new tensordict.

Parameters:

prefix (str) – directory prefix where the memory-mapped tensors will be stored. The directory tree structure will mimic the tensordict’s.
copy_existing (bool) – If False (default), an exception will be raised if an entry in the tensordict is already a tensor stored on disk with an associated file, but is not saved in the correct location according to prefix. If True, any existing Tensor will be copied to the new location.

Keyword Arguments:

num_threads (int, optional) – the number of threads used to write the memmap tensors. Defaults to 0.
return_early (bool, optional) – if True and num_threads>0, the method will return a future of the tensordict.
share_non_tensor (bool, optional) – if True, the non-tensor data will be shared between the processes and writing operation (such as inplace update or set) on any of the workers within a single node will update the value on all other workers. If the number of non-tensor leaves is high (e.g., sharing large stacks of non-tensor data) this may result in OOM or similar errors. Defaults to False.

The TensorDict is then locked, meaning that any writing operations that isn’t in-place will throw an exception (eg, rename, set or remove an entry). Once the tensordict is unlocked, the memory-mapped attribute is turned to False, because cross-process identity is not guaranteed anymore.

Returns:: A new tensordict with the tensors stored on disk if return_early=False, otherwise a TensorDictFuture instance.

Note

Serialising in this fashion might be slow with deeply nested tensordicts, so it is not recommended to call this method inside a training loop.

memmap_(prefix: str | None = None, copy_existing: bool = False, *, num_threads: int = 0, return_early: bool = False, share_non_tensor: bool = False) → T¶

Writes all tensors onto a corresponding memory-mapped Tensor, in-place.

Parameters:

prefix (str) – directory prefix where the memory-mapped tensors will be stored. The directory tree structure will mimic the tensordict’s.
copy_existing (bool) – If False (default), an exception will be raised if an entry in the tensordict is already a tensor stored on disk with an associated file, but is not saved in the correct location according to prefix. If True, any existing Tensor will be copied to the new location.

Keyword Arguments:

num_threads (int, optional) – the number of threads used to write the memmap tensors. Defaults to 0.
return_early (bool, optional) – if True and num_threads>0, the method will return a future of the tensordict.
share_non_tensor (bool, optional) – if True, the non-tensor data will be shared between the processes and writing operation (such as inplace update or set) on any of the workers within a single node will update the value on all other workers. If the number of non-tensor leaves is high (e.g., sharing large stacks of non-tensor data) this may result in OOM or similar errors. Defaults to False.

The TensorDict is then locked, meaning that any writing operations that isn’t in-place will throw an exception (eg, rename, set or remove an entry). Once the tensordict is unlocked, the memory-mapped attribute is turned to False, because cross-process identity is not guaranteed anymore.

Returns:: self if return_early=False, otherwise a TensorDictFuture instance.

Note

Serialising in this fashion might be slow with deeply nested tensordicts, so it is not recommended to call this method inside a training loop.

memmap_like(prefix: str | None = None, copy_existing: bool = False, *, num_threads: int = 0, return_early: bool = False, share_non_tensor: bool = False) → T¶

Creates a contentless Memory-mapped tensordict with the same shapes as the original one.

Parameters:

prefix (str) – directory prefix where the memory-mapped tensors will be stored. The directory tree structure will mimic the tensordict’s.
copy_existing (bool) – If False (default), an exception will be raised if an entry in the tensordict is already a tensor stored on disk with an associated file, but is not saved in the correct location according to prefix. If True, any existing Tensor will be copied to the new location.

Keyword Arguments:

num_threads (int, optional) – the number of threads used to write the memmap tensors. Defaults to 0.
return_early (bool, optional) – if True and num_threads>0, the method will return a future of the tensordict.
share_non_tensor (bool, optional) – if True, the non-tensor data will be shared between the processes and writing operation (such as inplace update or set) on any of the workers within a single node will update the value on all other workers. If the number of non-tensor leaves is high (e.g., sharing large stacks of non-tensor data) this may result in OOM or similar errors. Defaults to False.

The TensorDict is then locked, meaning that any writing operations that isn’t in-place will throw an exception (eg, rename, set or remove an entry). Once the tensordict is unlocked, the memory-mapped attribute is turned to False, because cross-process identity is not guaranteed anymore.

Returns:: A new TensorDict instance with data stored as memory-mapped tensors if return_early=False, otherwise a TensorDictFuture instance.

Note

This is the recommended method to write a set of large buffers on disk, as memmap_() will copy the information, which can be slow for large content.

Examples

>>> td = TensorDict({
...     "a": torch.zeros((3, 64, 64), dtype=torch.uint8),
...     "b": torch.zeros(1, dtype=torch.int64),
... }, batch_size=[]).expand(1_000_000)  # expand does not allocate new memory
>>> buffer = td.memmap_like("/path/to/dataset")

property names: Size¶

Retrieves the dim names for the tensor class.

Returns:: names (list of str)

set(key: Union[str, Tuple[str, ...]], value: Any, inplace: bool = False, non_blocking: bool = False)¶

Sets a new key-value pair.

Parameters:

key (str, tuple of str) – name of the key to be set. If tuple of str it is equivalent to chained calls of getattr followed by a final setattr.
value (Any) – value to be stored in the tensorclass
inplace (bool, optional) – if True, set will tentatively try to update the value in-place. If False or if the key isn’t present, the value will be simply written at its destination.

Returns:

self

state_dict(destination=None, prefix='', keep_vars=False, flatten=False) → dict[str, Any]¶: Returns a state_dict dictionary that can be used to save and load data from a tensorclass.

to_tensordict() → TensorDict¶

Convert the tensorclass into a regular TensorDict.

Makes a copy of all entries. Memmap and shared memory tensors are converted to regular tensors.

Returns:: A new TensorDict object containing the same values as the tensorclass.

unbind(dim: int)¶

Returns a tuple of indexed tensorclass instances unbound along the indicated dimension.

Resulting tensorclass instances will share the storage of the initial tensorclass instance.

RewardData¶

Docs

Tutorials

Resources