Serialization semantics ¶

This note describes how you can save and load PyTorch tensors and module states in Python, and how to serialize Python modules so they can be loaded in C++.

Table of Contents

Serialization semantics

Saving and loading tensors ¶

torch.save() and torch.load() let you easily save and load tensors:

>>> t = torch.tensor([1., 2.])
>>> torch.save(t, 'tensor.pt')
>>> torch.load('tensor.pt')
tensor([1., 2.])

By convention, PyTorch files are typically written with a ‘.pt’ or ‘.pth’ extension.

torch.save() and torch.load() use Python’s pickle by default, so you can also save multiple tensors as part of Python objects like tuples, lists, and dicts:

>>> d = {'a': torch.tensor([1., 2.]), 'b': torch.tensor([3., 4.])}
>>> torch.save(d, 'tensor_dict.pt')
>>> torch.load('tensor_dict.pt')
{'a': tensor([1., 2.]), 'b': tensor([3., 4.])}

Custom data structures that include PyTorch tensors can also be saved if the data structure is pickle-able.

Saving and loading tensors preserves views ¶

Saving tensors preserves their view relationships:

>>> numbers = torch.arange(1, 10)
>>> evens = numbers[1::2]
>>> torch.save([numbers, evens], 'tensors.pt')
>>> loaded_numbers, loaded_evens = torch.load('tensors.pt')
>>> loaded_evens *= 2
>>> loaded_numbers
tensor([ 1,  4,  3,  8,  5, 12,  7, 16,  9])

Behind the scenes, these tensors share the same “storage.” See Tensor Views for more on views and storage.

When PyTorch saves tensors it saves their storage objects and tensor metadata separately. This is an implementation detail that may change in the future, but it typically saves space and lets PyTorch easily reconstruct the view relationships between the loaded tensors. In the above snippet, for example, only a single storage is written to ‘tensors.pt’.

In some cases, however, saving the current storage objects may be unnecessary and create prohibitively large files. In the following snippet a storage much larger than the saved tensor is written to a file:

>>> large = torch.arange(1, 1000)
>>> small = large[0:5]
>>> torch.save(small, 'small.pt')
>>> loaded_small = torch.load('small.pt')
>>> loaded_small.storage().size()
999

Instead of saving only the five values in the small tensor to ‘small.pt,’ the 999 values in the storage it shares with large were saved and loaded.

When saving tensors with fewer elements than their storage objects, the size of the saved file can be reduced by first cloning the tensors. Cloning a tensor produces a new tensor with a new storage object containing only the values in the tensor:

>>> large = torch.arange(1, 1000)
>>> small = large[0:5]
>>> torch.save(small.clone(), 'small.pt')  # saves a clone of small
>>> loaded_small = torch.load('small.pt')
>>> loaded_small.storage().size()
5

Since the cloned tensors are independent of each other, however, they have none of the view relationships the original tensors did. If both file size and view relationships are important when saving tensors smaller than their storage objects, then care must be taken to construct new tensors that minimize the size of their storage objects but still have the desired view relationships before saving.

Saving and loading torch.nn.Modules ¶

In PyTorch, a module’s state is frequently serialized using a ‘state dict.’ A module’s state dict contains all of its parameters and persistent buffers:

>>> bn = torch.nn.BatchNorm1d(3, track_running_stats=True)
>>> list(bn.named_parameters())
[('weight', Parameter containing: tensor([1., 1., 1.], requires_grad=True)),
 ('bias', Parameter containing: tensor([0., 0., 0.], requires_grad=True))]

>>> list(bn.named_buffers())
[('running_mean', tensor([0., 0., 0.])),
 ('running_var', tensor([1., 1., 1.])),
 ('num_batches_tracked', tensor(0))]

>>> bn.state_dict()
OrderedDict([('weight', tensor([1., 1., 1.])),
             ('bias', tensor([0., 0., 0.])),
             ('running_mean', tensor([0., 0., 0.])),
             ('running_var', tensor([1., 1., 1.])),
             ('num_batches_tracked', tensor(0))])

Instead of saving a module directly, for compatibility reasons it is recommended to instead save only its state dict. Python modules even have a function, load_state_dict(), to restore their states from a state dict:

>>> torch.save(bn.state_dict(), 'bn.pt')
>>> bn_state_dict = torch.load('bn.pt')
>>> new_bn = torch.nn.BatchNorm1d(3, track_running_stats=True)
>>> new_bn.load_state_dict(bn_state_dict)
<All keys matched successfully>

Note that the state dict is first loaded from its file with torch.load() and the state then restored with load_state_dict().

Even custom modules and modules containing other modules have state dicts and can use this pattern:

# A module with two linear layers
>>> class MyModule(torch.nn.Module):
      def __init__(self):
        super().__init__()
        self.l0 = torch.nn.Linear(4, 2)
        self.l1 = torch.nn.Linear(2, 1)

      def forward(self, input):
        out0 = self.l0(input)
        out0_relu = torch.nn.functional.relu(out0)
        return self.l1(out0_relu)

>>> m = MyModule()
>>> m.state_dict()
OrderedDict([('l0.weight', tensor([[ 0.1400, 0.4563, -0.0271, -0.4406],
                                   [-0.3289, 0.2827, 0.4588, 0.2031]])),
             ('l0.bias', tensor([ 0.0300, -0.1316])),
             ('l1.weight', tensor([[0.6533, 0.3413]])),
             ('l1.bias', tensor([-0.1112]))])

>>> torch.save(m.state_dict(), 'mymodule.pt')
>>> m_state_dict = torch.load('mymodule.pt')
>>> new_m = MyModule()
>>> new_m.load_state_dict(m_state_dict)
<All keys matched successfully>

Serialized file format for `torch.save`¶

Since PyTorch 1.6.0, torch.save defaults to returning an uncompressed ZIP64 archive unless the user sets _use_new_zipfile_serialization=False.

In this archive, the files are ordered as such

checkpoint.pth
├── data.pkl
├── byteorder  # added in PyTorch 2.1.0
├── data/
│   ├── 0
│   ├── 1
│   ├── 2
│   └── …
└── version

The entries are as follows:

data.pkl is the result of pickling the object passed to torch.save excluding torch.Storage objects that it contains
byteorder contains a string with the sys.byteorder when saving (“little” or “big”)
data/ contains all the storages in the object, where each storage is a separate file
version contains a version number at save time that can be used at load time

When saving, PyTorch will ensure that the local file header of each file is padded to an offset that is a multiple of 64 bytes, ensuring that the offset of each file is 64-byte aligned.

Note

Tensors on certain devices such as XLA are serialized as pickled numpy arrays. As such, their storages are not serialized. In these cases data/ might not exist in the checkpoint.

`torch.load` with `weights_only=True`¶

Starting in version 2.6, torch.load will use weights_only=True if the pickle_module argument is not passed.

As discussed in the documentation for torch.load(), weights_only=True restricts the unpickler used in torch.load to only executing functions/building classes required for state_dicts of plain torch.Tensors as well as some other primitive types. Further, unlike the default Unpickler provided by the pickle module, the weights_only Unpickler is not allowed to dynamically import anything during unpickling.

As mentioned above, saving a module’s state_dict is a best practice when using torch.save. If loading an old checkpoint that contains an nn.Module, we recommend weights_only=False. When loading a checkpoint that contains tensor subclasses, there will likely be functions/classes that need to be allowlisted, see below for further details.

If the weights_only Unpickler encounters a function or class that is not allowlisted by default within the pickle file, you should see an actionable error like such

_pickle.UnpicklingError: Weights only load failed. This file can still be loaded,
to do so you have two options, do those steps only if you trust the source of the checkpoint.
    1. Re-running `torch.load` with `weights_only` set to `False` will likely succeed,
        but it can result in arbitrary code execution. Do it only if you got the file from a trusted source.
    2. Alternatively, to load with `weights_only=True` please check the recommended
       steps in the following error message.
       WeightsUnpickler error: Unsupported global: GLOBAL {__module__}.{__name__} was not an allowed global by
       default. Please use `torch.serialization.add_safe_globals([{__name__}])` or the
       `torch.serialization.safe_globals([{__name__}])` context manager to allowlist this global
       if you trust this class/function.

Please follow the steps in the error message and allowlist the functions or classes only if you trust them.

To get all GLOBALs (functions/classes) in the checkpoint that are not yet allowlisted you can use torch.serialization.get_unsafe_globals_in_checkpoint() which will return a list of strings of the form {__module__}.{__name__}. If you trust these functions/classes, you can import them and allowlist them per the error message either via torch.serialization.add_safe_globals() or the context manager torch.serialization.safe_globals.

To access the list of user-allowlisted functions/classes you can use torch.serialization.get_safe_globals() and to clear the current list see torch.serialization.clear_safe_globals().

Troubleshooting `weights_only`¶

Getting unsafe globals ¶

A caveat is that torch.serialization.get_unsafe_globals_in_checkpoint() analyzes the checkpoint statically, some types might be built dynamically during the unpickling process and hence will not be reported by torch.serialization.get_unsafe_globals_in_checkpoint(). One such example is dtypes in numpy. In numpy < 1.25 after allowlisting all the functions/classes reported by torch.serialization.get_unsafe_globals_in_checkpoint() you might see an error like

WeightsUnpickler error: Can only build Tensor, Parameter, OrderedDict or types allowlisted via `add_safe_globals`,
but got <class 'numpy.dtype[float32]'>

This can be allowlisted via {add_}safe_globals([type(np.dtype(np.float32))]).

In numpy >=1.25 you would see

WeightsUnpickler error: Can only build Tensor, Parameter, OrderedDict or types allowlisted via `add_safe_globals`,
but got <class 'numpy.dtypes.Float32DType'>

This can be allowlisted via {add_}safe_globals([np.dtypes.Float32DType]).

Environment Variables ¶

There are two environment variables that will influence the behavior of torch.load. These can be helpful if one does not have access to the torch.load callsites.

TORCH_FORCE_WEIGHTS_ONLY_LOAD=1 will override all torch.load callsites to use weights_only=True.
TORCH_FORCE_NO_WEIGHTS_ONLY_LOAD=1 will make torch.load callsites use weights_only=False only if weights_only was not passed as an argument.

Serializing torch.nn.Modules and loading them in C++¶

ScriptModules can be serialized as a TorchScript program and loaded using torch.jit.load(). This serialization encodes all the modules’ methods, submodules, parameters, and attributes, and it allows the serialized program to be loaded in C++ (i.e. without Python).

The distinction between torch.jit.save() and torch.save() may not be immediately clear. torch.save() saves Python objects with pickle. This is especially useful for prototyping, researching, and training. torch.jit.save(), on the other hand, serializes ScriptModules to a format that can be loaded in Python or C++. This is useful when saving and loading C++ modules or for running modules trained in Python with C++, a common practice when deploying PyTorch models.

To script, serialize and load a module in Python:

>>> scripted_module = torch.jit.script(MyModule())
>>> torch.jit.save(scripted_module, 'mymodule.pt')
>>> torch.jit.load('mymodule.pt')
RecursiveScriptModule( original_name=MyModule
                      (l0): RecursiveScriptModule(original_name=Linear)
                      (l1): RecursiveScriptModule(original_name=Linear) )

Traced modules can also be saved with torch.jit.save(), with the caveat that only the traced code path is serialized. The following example demonstrates this:

# A module with control flow
>>> class ControlFlowModule(torch.nn.Module):
      def __init__(self):
        super().__init__()
        self.l0 = torch.nn.Linear(4, 2)
        self.l1 = torch.nn.Linear(2, 1)

      def forward(self, input):
        if input.dim() > 1:
            return torch.tensor(0)

        out0 = self.l0(input)
        out0_relu = torch.nn.functional.relu(out0)
        return self.l1(out0_relu)

>>> traced_module = torch.jit.trace(ControlFlowModule(), torch.randn(4))
>>> torch.jit.save(traced_module, 'controlflowmodule_traced.pt')
>>> loaded = torch.jit.load('controlflowmodule_traced.pt')
>>> loaded(torch.randn(2, 4)))
tensor([[-0.1571], [-0.3793]], grad_fn=<AddBackward0>)

>>> scripted_module = torch.jit.script(ControlFlowModule(), torch.randn(4))
>>> torch.jit.save(scripted_module, 'controlflowmodule_scripted.pt')
>>> loaded = torch.jit.load('controlflowmodule_scripted.pt')
>> loaded(torch.randn(2, 4))
tensor(0)

The above module has an if statement that is not triggered by the traced inputs, and so is not part of the traced module and not serialized with it. The scripted module, however, contains the if statement and is serialized with it. See the TorchScript documentation for more on scripting and tracing.

Finally, to load the module in C++:

>>> torch::jit::script::Module module;
>>> module = torch::jit::load('controlflowmodule_scripted.pt');

See the PyTorch C++ API documentation for details about how to use PyTorch modules in C++.

Saving and loading ScriptModules across PyTorch versions ¶

The PyTorch Team recommends saving and loading modules with the same version of PyTorch. Older versions of PyTorch may not support newer modules, and newer versions may have removed or modified older behavior. These changes are explicitly described in PyTorch’s release notes, and modules relying on functionality that has changed may need to be updated to continue working properly. In limited cases, detailed below, PyTorch will preserve the historic behavior of serialized ScriptModules so they do not require an update.

torch.div performing integer division ¶

In PyTorch 1.5 and earlier torch.div() would perform floor division when given two integer inputs:

# PyTorch 1.5 (and earlier)
>>> a = torch.tensor(5)
>>> b = torch.tensor(3)
>>> a / b
tensor(1)

In PyTorch 1.7, however, torch.div() will always perform a true division of its inputs, just like division in Python 3:

# PyTorch 1.7
>>> a = torch.tensor(5)
>>> b = torch.tensor(3)
>>> a / b
tensor(1.6667)

The behavior of torch.div() is preserved in serialized ScriptModules. That is, ScriptModules serialized with versions of PyTorch before 1.6 will continue to see torch.div() perform floor division when given two integer inputs even when loaded with newer versions of PyTorch. ScriptModules using torch.div() and serialized on PyTorch 1.6 and later cannot be loaded in earlier versions of PyTorch, however, since those earlier versions do not understand the new behavior.

torch.full always inferring a float dtype ¶

In PyTorch 1.5 and earlier torch.full() always returned a float tensor, regardless of the fill value it’s given:

# PyTorch 1.5 and earlier
>>> torch.full((3,), 1)  # Note the integer fill value...
tensor([1., 1., 1.])     # ...but float tensor!

In PyTorch 1.7, however, torch.full() will infer the returned tensor’s dtype from the fill value:

# PyTorch 1.7
>>> torch.full((3,), 1)
tensor([1, 1, 1])

>>> torch.full((3,), True)
tensor([True, True, True])

>>> torch.full((3,), 1.)
tensor([1., 1., 1.])

>>> torch.full((3,), 1 + 1j)
tensor([1.+1.j, 1.+1.j, 1.+1.j])

The behavior of torch.full() is preserved in serialized ScriptModules. That is, ScriptModules serialized with versions of PyTorch before 1.6 will continue to see torch.full return float tensors by default, even when given bool or integer fill values. ScriptModules using torch.full() and serialized on PyTorch 1.6 and later cannot be loaded in earlier versions of PyTorch, however, since those earlier versions do not understand the new behavior.

Utility functions ¶

The following utility functions are related to serialization:

torch.serialization.register_package(priority, tagger, deserializer)[source][source]¶

Registers callables for tagging and deserializing storage objects with an associated priority. Tagging associates a device with a storage object at save time while deserializing moves a storage object to an appropriate device at load time. tagger and deserializer are run in the order given by their priority until a tagger/deserializer returns a value that is not None.

To override the deserialization behavior for a device in the global registry, one can register a tagger with a higher priority than the existing tagger.

This function can also be used to register a tagger and deserializer for new devices.

Parameters

priority (int) – Indicates the priority associated with the tagger and deserializer, where a lower value indicates higher priority.
tagger (Callable[[Union[Storage, TypedStorage, UntypedStorage]], Optional[str]]) – Callable that takes in a storage object and returns its tagged device as a string or None.
deserializer (Callable[[Union[Storage, TypedStorage, UntypedStorage], str], Optional[Union[Storage, TypedStorage, UntypedStorage]]]) – Callable that takes in storage object and a device string and returns a storage object on the appropriate device or None.

Returns

None

Example

>>> def ipu_tag(obj):
>>>     if obj.device.type == 'ipu':
>>>         return 'ipu'
>>> def ipu_deserialize(obj, location):
>>>     if location.startswith('ipu'):
>>>         ipu = getattr(torch, "ipu", None)
>>>         assert ipu is not None, "IPU device module is not loaded"
>>>         assert torch.ipu.is_available(), "ipu is not available"
>>>         return obj.ipu(location)
>>> torch.serialization.register_package(11, ipu_tag, ipu_deserialize)

torch.serialization.get_crc32_options()[source][source]¶

Get whether torch.save() computes and writes crc32 for each record.

Defaults to True.

Return type: bool

torch.serialization.set_crc32_options(compute_crc32)[source][source]¶

Set whether torch.save() computes and writes crc32 for each record.

Note

Setting this to False may make unzipping of the torch.save output fail or warn due to corrupted CRC32. However torch.load will be able to load the file.

Parameters: compute_crc32 (bool) – set crc32 compuation flag

torch.serialization.get_default_load_endianness()[source][source]¶

Get fallback byte order for loading files

If byteorder mark is not present in saved checkpoint, this byte order is used as fallback. By default, it’s “native” byte order.

Returns: Optional[LoadEndianness]
Return type: default_load_endian

torch.serialization.set_default_load_endianness(endianness)[source][source]¶

Set fallback byte order for loading files

If byteorder mark is not present in saved checkpoint, this byte order is used as fallback. By default, it’s “native” byte order.

Parameters: endianness – the new fallback byte order

torch.serialization.get_default_mmap_options()[source][source]¶

Get default mmap options for torch.load() with mmap=True.

Defaults to mmap.MAP_PRIVATE.

Returns: int
Return type: default_mmap_options

torch.serialization.set_default_mmap_options(flags)[source][source]¶

Context manager or function to set default mmap options for torch.load() with mmap=True to flags.

For now, only either mmap.MAP_PRIVATE or mmap.MAP_SHARED are supported. Please open an issue if you need any other option to be added here.

Note

This feature is currently not supported for Windows.

Parameters: flags (int) – mmap.MAP_PRIVATE or mmap.MAP_SHARED

torch.serialization.add_safe_globals(safe_globals)[source][source]¶

Marks the given globals as safe for weights_only load. For example, functions added to this list can be called during unpickling, classes could be instantiated and have state set.

Each item in the list can either be a function/class or a tuple of the form (function/class, string) where string is the full path of the function/class.

Within the serialized format, each function is identified with its full path as {__module__}.{__name__}. When calling this API, you can provide this full path that should match the one in the checkpoint otherwise the default {fn.__module__}.{fn.__name__} will be used.

Parameters: safe_globals (List[Union[Callable, Tuple[Callable, str]]]) – list of globals to mark as safe

Example

>>> import tempfile
>>> class MyTensor(torch.Tensor):
...     pass
>>> t = MyTensor(torch.randn(2, 3))
>>> with tempfile.NamedTemporaryFile() as f:
...     torch.save(t, f.name)
# Running `torch.load(f.name, weights_only=True)` will fail with
# Unsupported global: GLOBAL __main__.MyTensor was not an allowed global by default.
# Check the code and make sure MyTensor is safe to be used when loaded from an arbitrary checkpoint.
...     torch.serialization.add_safe_globals([MyTensor])
...     torch.load(f.name, weights_only=True)
# MyTensor([[-0.5024, -1.8152, -0.5455],
#          [-0.8234,  2.0500, -0.3657]])

torch.serialization.clear_safe_globals()[source][source]¶

Clears the list of globals that are safe for weights_only load.

torch.serialization.get_safe_globals()[source][source]¶

Returns the list of user-added globals that are safe for weights_only load.

Return type: List[Union[Callable, Tuple[Callable, str]]]

torch.serialization.get_unsafe_globals_in_checkpoint(f)[source][source]¶

Returns a list of strings of functions/classes in a torch.save object that are not safe for weights_only.

For a given function or class f, the corresponding string will be of the form {f.__module__}.{f.__name__}.

This function will return any GLOBALs in the checkpoint that are not in the set marked safe for weights_only (either via add_safe_globals() or safe_globals context or allowlisted by torch by default).

Note

This function will statically disassemble the pickle file in the checkpoint. The implication is any classes dynamically pushed onto the stack during unpickling will not be included in the output.

Parameters: f (Union[str, PathLike, BinaryIO, IO[bytes]]) – File-like object or string containing the checkpoint object saved via torch.save
Returns: A list of strings of pickle GLOBALs in the checkpoint that are not allowlisted for weights_only.
Return type: List[str]

class torch.serialization.safe_globals(safe_globals)[source][source]¶

Context-manager that adds certain globals as safe for weights_only load.

Parameters: safe_globals (List[Union[Callable, Tuple[Callable, str]]]) – List of globals for weights_only load.

Example

>>> import tempfile
>>> class MyTensor(torch.Tensor):
...     pass
>>> t = MyTensor(torch.randn(2, 3))
>>> with tempfile.NamedTemporaryFile() as f:
...     torch.save(t, f.name)
# Running `torch.load(f.name, weights_only=True)` will fail with
# Unsupported global: GLOBAL __main__.MyTensor was not an allowed global by default.
# Check the code and make sure MyTensor is safe to be used when loaded from an arbitrary checkpoint.
...     with torch.serialization.safe_globals([MyTensor]):
...         torch.load(f.name, weights_only=True)
# MyTensor([[-0.5024, -1.8152, -0.5455],
#          [-0.8234,  2.0500, -0.3657]])
>>> assert torch.serialization.get_safe_globals() == []

class torch.serialization.skip_data(materialize_fake_tensors=False)[source][source]¶

Context-manager that skips writing storage bytes for torch.save calls.

Storages will still be saved, but the space that their bytes would usually be written to will be empty space. The storage bytes can then be populated in a separate pass.

Warning

The skip_data context manager is an early prototype and is subject to change.

Parameters: materialize_fake_tensors (bool) – Whether to materialize FakeTensors.

Example

>>> import tempfile
>>> t = torch.randn(2, 3)
>>> with tempfile.NamedTemporaryFile() as f:
...     with torch.serialization.skip_data():
...         torch.save(t, f.name)
...     torch.load(f.name, weights_only=True)
tensor([[0., 0., 0.],
        [0., 0., 0.]])

Serialization semantics ¶

Saving and loading tensors ¶

Saving and loading tensors preserves views ¶

Saving and loading torch.nn.Modules ¶

Serialized file format for `torch.save`¶

`torch.load` with `weights_only=True`¶

Troubleshooting `weights_only`¶

Getting unsafe globals ¶

Environment Variables ¶

Serializing torch.nn.Modules and loading them in C++¶

Saving and loading ScriptModules across PyTorch versions ¶

torch.div performing integer division ¶

torch.full always inferring a float dtype ¶

Utility functions ¶

Docs

Tutorials

Resources