Skip to main content

No project description provided

Project description

Documentation Benchmarks Python version GitHub license pypi version pypi nightly version Downloads Downloads codecov pytorch

TensorDict

Installation | General features | Tensor-like features | Distributed capabilities | TensorDict for functional programming using FuncTorch | Lazy preallocation | Nesting TensorDicts | TensorClass

TensorDict is a dictionary-like class that inherits properties from tensors, such as indexing, shape operations, casting to device or point-to-point communication in distributed settings.

The main purpose of TensorDict is to make code-bases more readable and modular by abstracting away tailored operations:

for i, tensordict in enumerate(dataset):
    # the model reads and writes tensordicts
    tensordict = model(tensordict)
    loss = loss_module(tensordict)
    loss.backward()
    optimizer.step()
    optimizer.zero_grad()

With this level of abstraction, one can recycle a training loop for highly heterogeneous task. Each individual step of the training loop (data collection and transform, model prediction, loss computation etc.) can be tailored to the use case at hand without impacting the others. For instance, the above example can be easily used across classification and segmentation tasks, among many others.

Features

General

A tensordict is primarily defined by its batch_size (or shape) and its key-value pairs:

>>> from tensordict import TensorDict
>>> import torch
>>> tensordict = TensorDict({
...     "key 1": torch.ones(3, 4, 5),
...     "key 2": torch.zeros(3, 4, 5, dtype=torch.bool),
... }, batch_size=[3, 4])

The batch_size and the first dimensions of each of the tensors must be compliant. The tensors can be of any dtype and device. Optionally, one can restrict a tensordict to live on a dedicated device, which will send each tensor that is written there:

>>> tensordict = TensorDict({
...     "key 1": torch.ones(3, 4, 5),
...     "key 2": torch.zeros(3, 4, 5, dtype=torch.bool),
... }, batch_size=[3, 4], device="cuda:0")
>>> tensordict["key 3"] = torch.randn(3, 4, device="cpu")
>>> assert tensordict["key 3"].device is torch.device("cuda:0")

Tensor-like features

TensorDict objects can be indexed exactly like tensors. The resulting of indexing a TensorDict is another TensorDict containing tensors indexed along the required dimension:

>>> tensordict = TensorDict({
...     "key 1": torch.ones(3, 4, 5),
...     "key 2": torch.zeros(3, 4, 5, dtype=torch.bool),
... }, batch_size=[3, 4])
>>> sub_tensordict = tensordict[..., :2]
>>> assert sub_tensordict.shape == torch.Size([3, 2])
>>> assert sub_tensordict["key 1"].shape == torch.Size([3, 2, 5])

Similarly, one can build tensordicts by stacking or concatenating single tensordicts:

>>> tensordicts = [TensorDict({
...     "key 1": torch.ones(3, 4, 5),
...     "key 2": torch.zeros(3, 4, 5, dtype=torch.bool),
... }, batch_size=[3, 4]) for _ in range(2)]
>>> stack_tensordict = torch.stack(tensordicts, 1)
>>> assert stack_tensordict.shape == torch.Size([3, 2, 4])
>>> assert stack_tensordict["key 1"].shape == torch.Size([3, 2, 4, 5])
>>> cat_tensordict = torch.cat(tensordicts, 0)
>>> assert cat_tensordict.shape == torch.Size([6, 4])
>>> assert cat_tensordict["key 1"].shape == torch.Size([6, 4, 5])

TensorDict instances can also be reshaped, viewed, squeezed and unsqueezed:

>>> tensordict = TensorDict({
...     "key 1": torch.ones(3, 4, 5),
...     "key 2": torch.zeros(3, 4, 5, dtype=torch.bool),
... }, batch_size=[3, 4])
>>> print(tensordict.view(-1))
torch.Size([12])
>>> print(tensordict.reshape(-1))
torch.Size([12])
>>> print(tensordict.unsqueeze(-1))
torch.Size([3, 4, 1])

One can also send tensordict from device to device, place them in shared memory, clone them, update them in-place or not, split them, unbind them, expand them etc.

If a functionality is missing, it is easy to call it using apply() or apply_():

tensordict_uniform = tensordict.apply(lambda tensor: tensor.uniform_())

Distributed capabilities

Complex data structures can be cumbersome to synchronize in distributed settings. tensordict solves that problem with synchronous and asynchronous helper methods such as recv, irecv, send and isend that behave like their torch.distributed counterparts:

>>> # on all workers
>>> data = TensorDict({"a": torch.zeros(()), ("b", "c"): torch.ones(())}, [])
>>> # on worker 1
>>> data.isend(dst=0)
>>> # on worker 0
>>> data.irecv(src=1)

When nodes share a common scratch space, the MemmapTensor backend can be used to seamlessly send, receive and read a huge amount of data.

TensorDict for functional programming using FuncTorch

We also provide an API to use TensorDict in conjunction with FuncTorch. For instance, TensorDict makes it easy to concatenate model weights to do model ensembling:

>>> from torch import nn
>>> from tensordict import TensorDict
>>> from tensordict.nn import make_functional
>>> import torch
>>> from functorch import vmap
>>> layer1 = nn.Linear(3, 4)
>>> layer2 = nn.Linear(4, 4)
>>> model = nn.Sequential(layer1, layer2)
>>> # we represent the weights hierarchically
>>> weights1 = TensorDict(layer1.state_dict(), []).unflatten_keys(".")
>>> weights2 = TensorDict(layer2.state_dict(), []).unflatten_keys(".")
>>> params = make_functional(model)
>>> assert (params == TensorDict({"0": weights1, "1": weights2}, [])).all()
>>> # Let's use our functional module
>>> x = torch.randn(10, 3)
>>> out = model(x, params=params)  # params is the last arg (or kwarg)
>>> # an ensemble of models: we stack params along the first dimension...
>>> params_stack = torch.stack([params, params], 0)
>>> # ... and use it as an input we'd like to pass through the model
>>> y = vmap(model, (None, 0))(x, params_stack)
>>> print(y.shape)
torch.Size([2, 10, 4])

Moreover, tensordict modules are compatible with torch.fx and torch.compile, which means that you can get the best of both worlds: a codebase that is both readable and future-proof as well as efficient and portable!

Lazy preallocation

Pre-allocating tensors can be cumbersome and hard to scale if the list of preallocated items varies according to the script configuration. TensorDict solves this in an elegant way. Assume you are working with a function foo() -> TensorDict, e.g.

def foo():
    tensordict = TensorDict({}, batch_size=[])
    tensordict["a"] = torch.randn(3)
    tensordict["b"] = TensorDict({"c": torch.zeros(2)}, batch_size=[])
    return tensordict

and you would like to call this function repeatedly. You could do this in two ways. The first would simply be to stack the calls to the function:

tensordict = torch.stack([foo() for _ in range(N)])

However, you could also choose to preallocate the tensordict:

tensordict = TensorDict({}, batch_size=[N])
for i in range(N):
    tensordict[i] = foo()

which also results in a tensordict (when N = 10)

TensorDict(
    fields={
        a: Tensor(torch.Size([10, 3]), dtype=torch.float32),
        b: TensorDict(
            fields={
                c: Tensor(torch.Size([10, 2]), dtype=torch.float32)},
            batch_size=torch.Size([10]),
            device=None,
            is_shared=False)},
    batch_size=torch.Size([10]),
    device=None,
    is_shared=False)

When i==0, your empty tensordict will automatically be populated with empty tensors of batch-size N. After that, updates will be written in-place. Note that this would also work with a shuffled series of indices (pre-allocation does not require you to go through the tensordict in an ordered fashion).

Nesting TensorDicts

It is possible to nest tensordict. The only requirement is that the sub-tensordict should be indexable under the parent tensordict, i.e. its batch size should match (but could be longer than) the parent batch size.

We can switch easily between hierarchical and flat representations. For instance, the following code will result in a single-level tensordict with keys "key 1" and "key 2.sub-key":

>>> tensordict = TensorDict({
...     "key 1": torch.ones(3, 4, 5),
...     "key 2": TensorDict({"sub-key": torch.randn(3, 4, 5, 6)}, batch_size=[3, 4, 5])
... }, batch_size=[3, 4])
>>> tensordict_flatten = tensordict.flatten_keys(separator=".")

Accessing nested tensordicts can be achieved with a single index:

>>> sub_value = tensordict["key 2", "sub-key"]

TensorClass

Content flexibility comes at the cost of predictability. In some cases, developers may be looking for data structure with a more explicit behavior. tensordict provides a dataclass-like decorator that allows for the creation of custom dataclasses that support the tensordict operations:

>>> from tensordict.prototype import tensorclass
>>> import torch
>>>
>>> @tensorclass
... class MyData:
...    image: torch.Tensor
...    mask: torch.Tensor
...    label: torch.Tensor
...
...    def mask_image(self):
...        return self.image[self.mask.expand_as(self.image)].view(*self.batch_size, -1)
...
...    def select_label(self, label):
...        return self[self.label == label]
...
>>> images = torch.randn(100, 3, 64, 64)
>>> label = torch.randint(10, (100,))
>>> mask = torch.zeros(1, 64, 64, dtype=torch.bool).bernoulli_().expand(100, 1, 64, 64)
>>>
>>> data = MyData(images, mask, label=label, batch_size=[100])
>>>
>>> print(data.select_label(1))
MyData(
    image=Tensor(torch.Size([11, 3, 64, 64]), dtype=torch.float32),
    label=Tensor(torch.Size([11]), dtype=torch.int64),
    mask=Tensor(torch.Size([11, 1, 64, 64]), dtype=torch.bool),
    batch_size=torch.Size([11]),
    device=None,
    is_shared=False)
>>> print(data.mask_image().shape)
torch.Size([100, 6117])
>>> print(data.reshape(10, 10))
MyData(
    image=Tensor(torch.Size([10, 10, 3, 64, 64]), dtype=torch.float32),
    label=Tensor(torch.Size([10, 10]), dtype=torch.int64),
    mask=Tensor(torch.Size([10, 10, 1, 64, 64]), dtype=torch.bool),
    batch_size=torch.Size([10, 10]),
    device=None,
    is_shared=False)

As this example shows, one can write a specific data structures with dedicated methods while still enjoying the TensorDict artifacts such as shape operations (e.g. reshape or permutations), data manipulation (indexing, cat and stack) or calling arbitrary functions through the apply method (and many more).

Tensorclasses support nesting and, in fact, all the TensorDict features.

Installation

To install the latest stable version of tensordict, simply run

pip install tensordict

This will work with Python 3.7 and upward as well as PyTorch 1.12 and upward.

To enjoy the latest features, one can use

pip install tensordict-nightly

Citation

If you're using TensorDict, please refer to this BibTeX entry to cite this work:

@software{TensorDict,
  author = {Moens, Vincent},
  title = {{TensorDict: your PyTorch universal data carrier}},
  url = {https://github.com/pytorch-labs/tensordict},
  version = {0.1.0},
  year = {2023}
}

Disclaimer

TensorDict is at the beta-stage, meaning that there may be bc-breaking changes introduced, but they should come with a warranty. Hopefully these should not happen too often, as the current roadmap mostly involves adding new features and building compatibility with the broader PyTorch ecosystem.

License

TensorDict is licensed under the MIT License. See LICENSE for details.

Project details


Release history Release notifications | RSS feed

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distributions

No source distribution files available for this release.See tutorial on generating distribution archives.

Built Distributions

tensordict_nightly-2023.3.22-py310-none-any.whl (105.5 kB view details)

Uploaded Python 3.10

tensordict_nightly-2023.3.22-py39-none-any.whl (105.5 kB view details)

Uploaded Python 3.9

tensordict_nightly-2023.3.22-py38-none-any.whl (105.5 kB view details)

Uploaded Python 3.8

tensordict_nightly-2023.3.22-py37-none-any.whl (105.5 kB view details)

Uploaded Python 3.7

File details

Details for the file tensordict_nightly-2023.3.22-py310-none-any.whl.

File metadata

  • Download URL: tensordict_nightly-2023.3.22-py310-none-any.whl
  • Upload date:
  • Size: 105.5 kB
  • Tags: Python 3.10
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/1.15.0 pkginfo/1.8.3 requests/2.27.1 setuptools/44.1.1 requests-toolbelt/0.10.1 tqdm/4.64.1 CPython/2.7.17

File hashes

Hashes for tensordict_nightly-2023.3.22-py310-none-any.whl
Algorithm Hash digest
SHA256 698b1aba81a716148bf6d680a4180a5b65c671f253575dd3e3dbfd58e9f9aa44
MD5 c074131048c2d3a88bc689a45c4b0ff2
BLAKE2b-256 fb42c8adab8e6e38ca40ac203712c4beb487990027e28fa2b7c6b2e136f383d6

See more details on using hashes here.

File details

Details for the file tensordict_nightly-2023.3.22-py39-none-any.whl.

File metadata

  • Download URL: tensordict_nightly-2023.3.22-py39-none-any.whl
  • Upload date:
  • Size: 105.5 kB
  • Tags: Python 3.9
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/1.15.0 pkginfo/1.8.3 requests/2.27.1 setuptools/44.1.1 requests-toolbelt/0.10.1 tqdm/4.64.1 CPython/2.7.17

File hashes

Hashes for tensordict_nightly-2023.3.22-py39-none-any.whl
Algorithm Hash digest
SHA256 9e56ee19e78f7269c1e83bb06ef59b3eefed5325bc9eeb7f33864506dff4f0d5
MD5 db2e85347c29306da3b3bbeb98781f8f
BLAKE2b-256 729e4706d3d7ea0b866d9f477d41d49a35a142523c622f08be56b68753fdcf04

See more details on using hashes here.

File details

Details for the file tensordict_nightly-2023.3.22-py38-none-any.whl.

File metadata

  • Download URL: tensordict_nightly-2023.3.22-py38-none-any.whl
  • Upload date:
  • Size: 105.5 kB
  • Tags: Python 3.8
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/1.15.0 pkginfo/1.8.3 requests/2.27.1 setuptools/44.1.1 requests-toolbelt/0.10.1 tqdm/4.64.1 CPython/2.7.17

File hashes

Hashes for tensordict_nightly-2023.3.22-py38-none-any.whl
Algorithm Hash digest
SHA256 3018c3f9d99adf5484e65817b71abf6d07100cf8ef16688416581ce431fff185
MD5 2000f957a9fff8699c221a291b119333
BLAKE2b-256 821b92eef9813caccc0ed4385f1db2ac9a1585627404fc34b80e4d7ae948c2ce

See more details on using hashes here.

File details

Details for the file tensordict_nightly-2023.3.22-py37-none-any.whl.

File metadata

  • Download URL: tensordict_nightly-2023.3.22-py37-none-any.whl
  • Upload date:
  • Size: 105.5 kB
  • Tags: Python 3.7
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/1.15.0 pkginfo/1.8.3 requests/2.27.1 setuptools/44.1.1 requests-toolbelt/0.10.1 tqdm/4.64.1 CPython/2.7.17

File hashes

Hashes for tensordict_nightly-2023.3.22-py37-none-any.whl
Algorithm Hash digest
SHA256 567c1bceb0d5896a3f779c4ea05c0e0e06364ef639f8aabbb9f6a8d92fc1e18e
MD5 a828ab88fe297856f336df0d5e27d74d
BLAKE2b-256 1be897629b040ff931ef1421990d31d989e19a8d3670ffcf129a6505f28c8573

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page