.. automodule:: torch.package
.. py:module:: torch.package.analyze
.. currentmodule:: torch.package
torch.package
adds support for creating packages containing both artifacts and arbitrary
PyTorch code. These packages can be saved, shared, used to load and execute models
at a later date or on a different machine, and can even be deployed to production using
torch::deploy
.
This document contains tutorials, how-to guides, explanations, and an API reference that
will help you learn more about torch.package
and how to use it.
Warning
This module depends on the pickle
module which is not secure. Only unpackage data you trust.
It is possible to construct malicious pickle data which will execute arbitrary code during unpickling. Never unpackage data that could have come from an untrusted source, or that could have been tampered with.
For more information, review the documentation for the pickle
module.
- Tutorials
- How do I...
- See what is inside a package?
- See why a given module was included as a dependency?
- Include arbitrary resources with my package and access them later?
- Customize how a class is packaged?
- Test in my source code whether or not it is executing inside a package?
- Patch code into a package?
- Access package contents from packaged code?
- Distinguish between packaged code and non-packaged code?
- Re-export an imported object?
- Package a TorchScript module?
- Explanation
- API Reference
A tutorial that guides you through packaging and unpackaging a simple model is available on Colab. After completing this exercise, you will be familiar with the basic API for creating and using Torch packages.
The container format for a torch.package
is ZIP, so any tools that work with standard ZIP files should
work for exploring the contents. Some common ways to interact with ZIP files:
unzip my_package.pt
will unzip thetorch.package
archive to disk, where you can freely inspect its contents.
$ unzip my_package.pt && tree my_package my_package ├── .data │ ├── 94304870911616.storage │ ├── 94304900784016.storage │ ├── extern_modules │ └── version ├── models │ └── model_1.pkl └── torchvision └── models ├── resnet.py └── utils.py ~ cd my_package && cat torchvision/models/resnet.py ...
- The Python
zipfile
module provides a standard way to read and write ZIP archive contents.
from zipfile import ZipFile with ZipFile("my_package.pt") as myzip: file_bytes = myzip.read("torchvision/models/resnet.py") # edit file_bytes in some way myzip.writestr("torchvision/models/resnet.py", new_file_bytes)
- vim has the ability to natively read ZIP archives. You can even edit files and :
write
them back into the archive!
# add this to your .vimrc to treat `*.pt` files as zip files au BufReadCmd *.pt call zip#Browse(expand("<amatch>")) ~ vi my_package.pt
:class:`PackageImporter` provides a file_structure()
method, which will return a printable
and queryable :class:`Directory` object. The :class:`Directory` object is a simple directory structure that you can use to explore the
current contents of a torch.package
.
The :class:`Directory` object itself is directly printable and will print out a file tree representation. To filter what is returned,
use the glob-style include
and exclude
filtering arguments.
with PackageExporter('my_package.pt') as pe: pe.save_pickle('models', 'model_1.pkl', mod) importer = PackageImporter('my_package.pt') # can limit printed items with include/exclude args print(importer.file_structure(include=["**/utils.py", "**/*.pkl"], exclude="**/*.storage")) print(importer.file_structure()) # will print out all files
Output:
# filtered with glob pattern: # include=["**/utils.py", "**/*.pkl"], exclude="**/*.storage" ─── my_package.pt ├── models │ └── model_1.pkl └── torchvision └── models └── utils.py # all files ─── my_package.pt ├── .data │ ├── 94304870911616.storage │ ├── 94304900784016.storage │ ├── extern_modules │ └── version ├── models │ └── model_1.pkl └── torchvision └── models ├── resnet.py └── utils.py
You can also query :class:`Directory` objects with the has_file()
method.
importer_file_structure = importer.file_structure() found: bool = importer_file_structure.has_file("package_a/subpackage.py")
Say there is a given module foo
, and you want to know why your :class:`PackageExporter` is pulling in foo
as a dependency.
:meth:`PackageExporter.get_rdeps` will return all modules that directly depend on foo
.
If you would like to see how a given module src
depends on foo
, the :meth:`PackageExporter.all_paths` method will
return a DOT-formatted graph showing all the dependency paths between src
and foo
.
If you would just like to see the whole dependency graph of your :class:`PackageExporter`, you can use :meth:`PackageExporter.dependency_graph_string`.
:class:`PackageExporter` exposes three methods, save_pickle
, save_text
and save_binary
that allow you to save
Python objects, text, and binary data to a package.
with torch.PackageExporter("package.pt") as exporter: # Pickles the object and saves to `my_resources/tensor.pkl` in the archive. exporter.save_pickle("my_resources", "tensor.pkl", torch.randn(4)) exporter.save_text("config_stuff", "words.txt", "a sample string") exporter.save_binary("raw_data", "binary", my_bytes)
:class:`PackageImporter` exposes complementary methods named load_pickle
, load_text
and load_binary
that allow you to load
Python objects, text and binary data from a package.
importer = torch.PackageImporter("package.pt") my_tensor = importer.load_pickle("my_resources", "tensor.pkl") text = importer.load_text("config_stuff", "words.txt") binary = importer.load_binary("raw_data", "binary")
torch.package
allows for the customization of how classes are packaged. This behavior is accessed through defining the method
__reduce_package__
on a class and by defining a corresponding de-packaging function. This is similar to defining __reduce__
for
Python’s normal pickling process.
Steps:
- Define the method
__reduce_package__(self, exporter: PackageExporter)
on the target class. This method should do the work to save the class instance inside of the package, and should return a tuple of the corresponding de-packaging function with the arguments needed to invoke the de-packaging function. This method is called by thePackageExporter
when it encounters an instance of the target class. - Define a de-packaging function for the class. This de-packaging function should do the work to reconstruct and return an instance of the class. The function signature’s first parameter should be a
PackageImporter
instance, and the rest of the parameters are user defined.
# foo.py [Example of customizing how class Foo is packaged] from torch.package import PackageExporter, PackageImporter import time class Foo: def __init__(self, my_string: str): super().__init__() self.my_string = my_string self.time_imported = 0 self.time_exported = 0 def __reduce_package__(self, exporter: PackageExporter): """ Called by ``torch.package.PackageExporter``'s Pickler's ``persistent_id`` when saving an instance of this object. This method should do the work to save this object inside of the ``torch.package`` archive. Returns function w/ arguments to load the object from a ``torch.package.PackageImporter``'s Pickler's ``persistent_load`` function. """ # use this pattern to ensure no naming conflicts with normal dependencies, # anything saved under this module name shouldn't conflict with other # items in the package generated_module_name = f"foo-generated._{exporter.get_unique_id()}" exporter.save_text( generated_module_name, "foo.txt", self.my_string + ", with exporter modification!", ) time_exported = time.clock_gettime(1) # returns de-packaging function w/ arguments to invoke with return (unpackage_foo, (generated_module_name, time_exported,)) def unpackage_foo( importer: PackageImporter, generated_module_name: str, time_exported: float ) -> Foo: """ Called by ``torch.package.PackageImporter``'s Pickler's ``persistent_load`` function when depickling a Foo object. Performs work of loading and returning a Foo instance from a ``torch.package`` archive. """ time_imported = time.clock_gettime(1) foo = Foo(importer.load_text(generated_module_name, "foo.txt")) foo.time_imported = time_imported foo.time_exported = time_exported return foo
# example of saving instances of class Foo import torch from torch.package import PackageImporter, PackageExporter import foo foo_1 = foo.Foo("foo_1 initial string") foo_2 = foo.Foo("foo_2 initial string") with PackageExporter('foo_package.pt') as pe: # save as normal, no extra work necessary pe.save_pickle('foo_collection', 'foo1.pkl', foo_1) pe.save_pickle('foo_collection', 'foo2.pkl', foo_2) pi = PackageImporter('foo_package.pt') print(pi.file_structure()) imported_foo = pi.load_pickle('foo_collection', 'foo1.pkl') print(f"foo_1 string: '{imported_foo.my_string}'") print(f"foo_1 export time: {imported_foo.time_exported}") print(f"foo_1 import time: {imported_foo.time_imported}")
# output of running above script ─── foo_package ├── foo-generated │ ├── _0 │ │ └── foo.txt │ └── _1 │ └── foo.txt ├── foo_collection │ ├── foo1.pkl │ └── foo2.pkl └── foo.py foo_1 string: 'foo_1 initial string, with reduction modification!' foo_1 export time: 9857706.650140837 foo_1 import time: 9857706.652698385
A :class:`PackageImporter` will add the attribute __torch_package__
to every module that it initializes. Your code can check for the
presence of this attribute to determine whether it is executing in a packaged context or not.
# In foo/bar.py: if "__torch_package__" in dir(): # true if the code is being loaded from a package def is_in_package(): return True UserException = Exception else: def is_in_package(): return False UserException = UnpackageableException
Now, the code will behave differently depending on whether it’s imported normally through your Python environment or imported from a
torch.package
.
from foo.bar import is_in_package print(is_in_package()) # False loaded_module = PackageImporter(my_package).import_module("foo.bar") loaded_module.is_in_package() # True
Warning: in general, it’s bad practice to have code that behaves differently depending on whether it’s packaged or not. This can lead to hard-to-debug issues that are sensitive to how you imported your code. If your package is intended to be heavily used, consider restructuring your code so that it behaves the same way no matter how it was loaded.
:class:`PackageExporter` offers a save_source_string()
method that allows one to save arbitrary Python source code to a module of your choosing.
with PackageExporter(f) as exporter: # Save the my_module.foo available in your current Python environment. exporter.save_module("my_module.foo") # This saves the provided string to my_module/foo.py in the package archive. # It will override the my_module.foo that was previously saved. exporter.save_source_string("my_module.foo", textwrap.dedent( """\ def my_function(): print('hello world') """ )) # If you want to treat my_module.bar as a package # (e.g. save to `my_module/bar/__init__.py` instead of `my_module/bar.py) # pass is_package=True, exporter.save_source_string("my_module.bar", "def foo(): print('hello')\n", is_package=True) importer = PackageImporter(f) importer.import_module("my_module.foo").my_function() # prints 'hello world'
:class:`PackageImporter` implements the importlib.resources API for accessing resources from inside a package.
with PackageExporter(f) as exporter: # saves text to my_resource/a.txt in the archive exporter.save_text("my_resource", "a.txt", "hello world!") # saves the tensor to my_pickle/obj.pkl exporter.save_pickle("my_pickle", "obj.pkl", torch.ones(2, 2)) # see below for module contents exporter.save_module("foo") exporter.save_module("bar")
The importlib.resources
API allows access to resources from within packaged code.
# foo.py: import importlib.resources import my_resource # returns "hello world!" def get_my_resource(): return importlib.resources.read_text(my_resource, "a.txt")
Using importlib.resources
is the recommended way to access package contents from within packaged code, since it complies
with the Python standard. However, it is also possible to access the parent :class:`PackageImporter` instance itself from within
packaged code.
# bar.py: import torch_package_importer # this is the PackageImporter that imported this module. # Prints "hello world!", equivalent to importlib.resources.read_text def get_my_resource(): return torch_package_importer.load_text("my_resource", "a.txt") # You also do things that the importlib.resources API does not support, like loading # a pickled object from the package. def get_my_pickle(): return torch_package_importer.load_pickle("my_pickle", "obj.pkl")
To tell if an object’s code is from a torch.package
, use the torch.package.is_from_package()
function.
Note: if an object is from a package but its definition is from a module marked extern
or from stdlib
,
this check will return False
.
importer = PackageImporter(f) mod = importer.import_module('foo') obj = importer.load_pickle('model', 'model.pkl') txt = importer.load_text('text', 'my_test.txt') assert is_from_package(mod) assert is_from_package(obj) assert not is_from_package(txt) # str is from stdlib, so this will return False
To re-export an object that was previously imported by a :class:`PackageImporter`, you must make the new :class:`PackageExporter` aware of the original :class:`PackageImporter` so that it can find source code for your object’s dependencies.
importer = PackageImporter(f) obj = importer.load_pickle("model", "model.pkl") # re-export obj in a new package with PackageExporter(f2, importer=(importer, sys_importer)) as exporter: exporter.save_pickle("model", "model.pkl", obj)
To package a TorchScript model, use the same save_pickle
and load_pickle
APIs as you would with any other object.
Saving TorchScript objects that are attributes or submodules is supported as well with no extra work.
# save TorchScript just like any other object with PackageExporter(file_name) as e: e.save_pickle("res", "script_model.pkl", scripted_model) e.save_pickle("res", "mixed_model.pkl", python_model_with_scripted_submodule) # load as normal importer = PackageImporter(file_name) loaded_script = importer.load_pickle("res", "script_model.pkl") loaded_mixed = importer.load_pickle("res", "mixed_model.pkl"
A torch.package
file is a ZIP archive which conventionally uses the .pt
extension. Inside the ZIP archive, there are two kinds of files:
- Framework files, which are placed in the
.data/
. - User files, which is everything else.
As an example, this is what a fully packaged ResNet model from torchvision
looks like:
resnet ├── .data # All framework-specific data is stored here. │ │ # It's named to avoid conflicts with user-serialized code. │ ├── 94286146172688.storage # tensor data │ ├── 94286146172784.storage │ ├── extern_modules # text file with names of extern modules (e.g. 'torch') │ ├── version # version metadata │ ├── ... ├── model # the pickled model │ └── model.pkl └── torchvision # all code dependencies are captured as source files └── models ├── resnet.py └── utils.py
The .data/
directory is owned by torch.package, and its contents are considered to be a private implementation detail.
The torch.package
format makes no guarantees about the contents of .data/
, but any changes made will be backward compatible
(that is, newer version of PyTorch will always be able to load older torch.packages
).
Currently, the .data/
directory contains the following items:
version
: a version number for the serialized format, so that thetorch.package
import infrastructures knows how to load this package.extern_modules
: a list of modules that are consideredextern:class:`PackageImporter`. ``extern
modules will be imported using the loading environment’s system importer.*.storage
: serialized tensor data.
.data ├── 94286146172688.storage ├── 94286146172784.storage ├── extern_modules ├── version ├── ...
All other files in the archive were put there by a user. The layout is identical to a Python regular package. For a deeper dive in how Python packaging works, please consult this essay (it’s slightly out of date, so double-check implementation details with the Python reference documentation).
<package root> ├── model # the pickled model │ └── model.pkl ├── another_package │ ├── __init__.py │ ├── foo.txt # a resource file , see importlib.resources │ └── ... └── torchvision └── models ├── resnet.py # torchvision.models.resnet └── utils.py # torchvision.models.utils
When you issue a save_pickle(obj, ...)
call, :class:`PackageExporter` will pickle the object normally. Then, it uses the
pickletools
standard library module to parse the pickle bytecode.
In a pickle, an object is saved along with a GLOBAL
opcode that describes where to find the implementation of the object’s type, like:
GLOBAL 'torchvision.models.resnet Resnet`
The dependency resolver will gather up all GLOBAL
ops and mark them as dependencies of your pickled object.
For more information about pickling and the pickle format, please consult the Python docs.
When a Python module is identified as a dependency, torch.package
walks the module’s python AST representation and looks for import statements with
full support for the standard forms: from x import y
, import z
, from w import v as u
, etc. When one of these import statements are
encountered, torch.package
registers the imported modules as dependencies that are then themselves parsed in the same AST walking way.
Note: AST parsing has limited support for the __import__(...)
syntax and does not support importlib.import_module
calls. In general, you should
not expect dynamic imports to be detected by torch.package
.
torch.package
automatically finds the Python modules that your code and objects depend on. This process is called dependency resolution.
For each module that the dependency resolver finds, you must specify an action to take.
The allowed actions are:
intern
: put this module into the package.extern
: declare this module as an external dependency of the package.mock
: stub out this module.deny
: depending on this module will raise an error during package export.
Finally, there is one more important action that is not technically part of torch.package
:
- Refactoring: remove or change the dependencies in your code.
Note that actions are only defined on entire Python modules. There is no way to package “just” a function or class from a module and leave the rest out.
This is by design. Python does not offer clean boundaries between objects defined in a module. The only defined unit of dependency organization is a
module, so that’s what torch.package
uses.
Actions are applied to modules using patterns. Patterns can either be module names ("foo.bar"
) or globs (like "foo.**"
). You associate a pattern
with an action using methods on :class:`PackageExporter`, e.g.
my_exporter.intern("torchvision.**") my_exporter.extern("numpy")
If a module matches a pattern, the corresponding action is applied to it. For a given module, patterns will be checked in the order that they were defined, and the first action will be taken.
If a module is intern
-ed, it will be placed into the package.
This action is your model code, or any related code you want to package. For example, if you are trying to package a ResNet from torchvision
,
you will need to intern
the module torchvision.models.resnet.
On package import, when your packaged code tries to import an intern
-ed module, PackageImporter will look inside your package for that module.
If it can’t find that module, an error will be raised. This ensures that each :class:`PackageImporter` is isolated from the loading environment—even
if you have my_interned_module
available in both your package and the loading environment, :class:`PackageImporter` will only use the version in your
package.
Note: Only Python source modules can be intern
-ed. Other kinds of modules, like C extension modules and bytecode modules, will raise an error if
you attempt to intern
them. These kinds of modules need to be mock
-ed or extern
-ed.
If a module is extern
-ed, it will not be packaged. Instead, it will be added to a list of external dependencies for this package. You can find this
list on package_exporter.extern_modules
.
On package import, when the packaged code tries to import an extern
-ed module, :class:`PackageImporter` will use the default Python importer to find
that module, as if you did importlib.import_module("my_externed_module")
. If it can’t find that module, an error will be raised.
In this way, you can depend on third-party libraries like numpy
and scipy
from within your package without having to package them too.
Warning: If any external library changes in a backwards-incompatible way, your package may fail to load. If you need long-term reproducibility
for your package, try to limit your use of extern
.
If a module is mock
-ed, it will not be packaged. Instead a stub module will be packaged in its place. The stub module will allow you to retrieve
objects from it (so that from my_mocked_module import foo
will not error), but any use of that object will raise a NotImplementedError
.
mock
should be used for code that you “know” will not be needed in the loaded package, but you still want available for use in non-packaged contents.
For example, initialization/configuration code, or code only used for debugging/training.
Warning: In general, mock
should be used as a last resort. It introduces behavioral differences between packaged code and non-packaged code,
which may lead to later confusion. Prefer instead to refactor your code to remove unwanted dependencies.
The best way to manage dependencies is to not have dependencies at all! Often, code can be refactored to remove unnecessary dependencies. Here are some guidelines for writing code with clean dependencies (which are also generally good practices!):
Include only what you use. Do not leave unused imports in your code. The dependency resolver is not smart enough to tell that they are indeed unused, and will try to process them.
Qualify your imports. For example, instead of writing import foo and later using foo.bar.baz
, prefer to write from foo.bar import baz
. This more
precisely specifies your real dependency (foo.bar
) and lets the dependency resolver know you don’t need all of foo
.
Split up large files with unrelated functionality into smaller ones. If your utils
module contains a hodge-podge of unrelated functionality, any module
that depends on utils
will need to pull in lots of unrelated dependencies, even if you only needed a small part of it. Prefer instead to define
single-purpose modules that can be packaged independently of one another.
Patterns allow you to specify groups of modules with a convenient syntax. The syntax and behavior of patterns follows the Bazel/Buck glob().
A module that we are trying to match against a pattern is called a candidate. A candidate is composed of a list of segments separated by a
separator string, e.g. foo.bar.baz
.
A pattern contains one or more segments. Segments can be:
- A literal string (e.g.
foo
), which matches exactly. - A string containing a wildcard (e.g.
torch
, orfoo*baz*
). The wildcard matches any string, including the empty string. - A double wildcard (
**
). This matches against zero or more complete segments.
Examples:
torch.**
: matchestorch
and all its submodules, e.g.torch.nn
andtorch.nn.functional
.torch.*
: matchestorch.nn
ortorch.functional
, but nottorch.nn.functional
ortorch
torch*.**
: matchestorch
,torchvision
, and all of their submodules
When specifying actions, you can pass multiple patterns, e.g.
exporter.intern(["torchvision.models.**", "torchvision.utils.**"])
A module will match against this action if it matches any of the patterns.
You can also specify patterns to exclude, e.g.
exporter.mock("**", exclude=["torchvision.**"])
A module will not match against this action if it matches any of the exclude patterns. In this example, we are mocking all modules except
torchvision
and its submodules.
When a module could potentially match against multiple actions, the first action defined will be taken.
Python makes it really easy to bind objects and run code at module-level scope. This is generally fine—after all, functions and classes are bound to names this way. However, things become more complicated when you define an object at module scope with the intention of mutating it, introducing mutable global state.
Mutable global state is quite useful—it can reduce boilerplate, allow for open registration into tables, etc. But unless employed very carefully, it can
cause complications when used with torch.package
.
Every :class:`PackageImporter` creates an independent environment for its contents. This is nice because it means we load multiple packages and ensure they are isolated from each other, but when modules are written in a way that assumes shared mutable global state, this behavior can create hard-to-debug errors.
Any class that you import from a :class:`PackageImporter` will be a version of the class specific to that importer. For example:
from foo import MyClass my_class_instance = MyClass() with PackageExporter(f) as exporter: exporter.save_module("foo") importer = PackageImporter(f) imported_MyClass = importer.import_module("foo").MyClass assert isinstance(my_class_instance, MyClass) # works assert isinstance(my_class_instance, imported_MyClass) # ERROR!
In this example, MyClass
and imported_MyClass
are not the same type. In this specific example, MyClass
and imported_MyClass
have exactly the
same implementation, so you might think it’s okay to consider them the same class. But consider the situation where imported_MyClass
is coming from an
older package with an entirely different implementation of MyClass
— in that case, it’s unsafe to consider them the same class.
Under the hood, each importer has a prefix that allows it to uniquely identify classes:
print(MyClass.__name__) # prints "foo.MyClass" print(imported_MyClass.__name__) # prints <torch_package_0>.foo.MyClass
That means you should not expect isinstance
checks to work when one of the arguments is from a package and the other is not. If you need this
functionality, consider the following options:
- Doing duck typing (just using the class instead of explicitly checking that it is of a given type).
- Make the typing relationship an explicit part of the class contract. For example, you can add an attribute tag
self.handler = "handle_me_this_way"
and have client code check for the value ofhandler
instead of checking the type directly.
Each :class:`PackageImporter` instance creates an independent, isolated environment for its modules and objects. Modules in a package can only import
other packaged modules, or modules marked extern
. If you use multiple :class:`PackageImporter` instances to load a single package, you will get
multiple independent environments that do not interact.
This is achieved by extending Python’s import infrastructure with a custom importer. :class:`PackageImporter` provides the same core API as the
importlib
importer; namely, it implements the import_module
and __import__
methods.
When you invoke :meth:`PackageImporter.import_module`, :class:`PackageImporter` will construct and return a new module, much as the system importer does.
However, :class:`PackageImporter` patches the returned module to use self
(i.e. that :class:`PackageImporter` instance) to fulfill future import
requests by looking in the package rather than searching the user’s Python environment.
To avoid confusion (“is this foo.bar
object the one from my package, or the one from my Python environment?”), :class:`PackageImporter` mangles the
__name__
and __file__
of all imported modules, by adding a mangle prefix to them.
For __name__
, a name like torchvision.models.resnet18
becomes <torch_package_0>.torchvision.models.resnet18
.
For __file__
, a name like torchvision/models/resnet18.py
becomes <torch_package_0>.torchvision/modules/resnet18.py
.
Name mangling helps avoid inadvertent punning of module names between different packages, and helps you debug by making stack traces and print
statements more clearly show whether they are referring to packaged code or not. For developer-facing details about mangling, consult
mangling.md
in torch/package/
.
.. autoclass:: torch.package.PackagingError
.. autoclass:: torch.package.EmptyMatchError
.. autoclass:: torch.package.PackageExporter :members: .. automethod:: __init__
.. autoclass:: torch.package.PackageImporter :members: .. automethod:: __init__
.. autoclass:: torch.package.Directory :members: