Posit AI Weblog: safetensors 0.1.0

May 25, 2024

safetensors is a brand new, easy, quick, and protected file format for storing tensors. The design of the file format and its authentic implementation are being led
by Hugging Face, and it’s getting largely adopted of their common ‘transformers’ framework. The safetensors R package deal is a pure-R implementation, permitting to each learn and write safetensor information.

The preliminary model (0.1.0) of safetensors is now on CRAN.

Motivation

The principle motivation for safetensors within the Python neighborhood is safety. As famous
within the official documentation:

The principle rationale for this crate is to take away the necessity to use pickle on PyTorch which is utilized by default.

Pickle is taken into account an unsafe format, because the motion of loading a Pickle file can
set off the execution of arbitrary code. This has by no means been a priority for torch
for R customers, for the reason that Pickle parser that’s included in LibTorch solely helps a subset
of the Pickle format, which doesn’t embody executing code.

Nevertheless, the file format has further benefits over different generally used codecs, together with:

Help for lazy loading: You’ll be able to select to learn a subset of the tensors saved within the file.
Zero copy: Studying the file doesn’t require extra reminiscence than the file itself.
(Technically the present R implementation does makes a single copy, however that may
be optimized out if we actually want it in some unspecified time in the future).
Easy: Implementing the file format is easy, and doesn’t require complicated dependencies.
Which means that it’s format for exchanging tensors between ML frameworks and
between totally different programming languages. For example, you possibly can write a safetensors file
in R and cargo it in Python, and vice-versa.

There are further benefits in comparison with different file codecs frequent on this house, and
you possibly can see a comparability desk right here.

Format

The safetensors format is described within the determine beneath. It’s principally a header file
containing some metadata, adopted by uncooked tensor buffers.

Primary utilization

safetensors may be put in from CRAN utilizing:

set up.packages("safetensors")

library(torch)
library(safetensors)

tensors <- listing(
  x = torch_randn(10, 10),
  y = torch_ones(10, 10)
)

str(tensors)
#> Record of two
#>  $ x:Float [1:10, 1:10]
#>  $ y:Float [1:10, 1:10]

tmp <- tempfile()
safe_save_file(tensors, tmp)

tensors <- safe_load_file(tmp)
str(tensors)
#> Record of two
#>  $ x:Float [1:10, 1:10]
#>  $ y:Float [1:10, 1:10]
#>  - attr(*, "metadata")=Record of two
#>   ..$ x:Record of three
#>   .. ..$ form       : int [1:2] 10 10
#>   .. ..$ dtype       : chr "F32"
#>   .. ..$ data_offsets: int [1:2] 0 400
#>   ..$ y:Record of three
#>   .. ..$ form       : int [1:2] 10 10
#>   .. ..$ dtype       : chr "F32"
#>   .. ..$ data_offsets: int [1:2] 400 800
#>  - attr(*, "max_offset")= int 929

At present, safetensors solely helps writing torch tensors, however we plan so as to add
assist for writing plain R arrays and tensorflow tensors sooner or later.

Future instructions

The subsequent model of torch will use safetensors as its serialization format,
that means that when calling torch_save() on a mannequin, listing of tensors, or different
sorts of objects supported by torch_save, you’re going to get a sound safetensors file.

That is an enchancment over the earlier implementation as a result of:

It’s a lot quicker. Greater than 10x for medium sized fashions. Might be much more for big information.
This additionally improves the efficiency of parallel dataloaders by ~30%.
It enhances cross-language and cross-framework compatibility. You’ll be able to prepare your mannequin
in R and use it in Python (and vice-versa), or prepare your mannequin in tensorflow and run it
with torch.

If you wish to strive it out, you possibly can set up the event model of torch with:

remotes::install_github("mlverse/torch")

Nick Fewings on Unsplash

Reuse

Textual content and figures are licensed beneath Inventive Commons Attribution CC BY 4.0. The figures which were reused from different sources do not fall beneath this license and may be acknowledged by a word of their caption: “Determine from …”.

Quotation

For attribution, please cite this work as

Falbel (2023, June 15). Posit AI Weblog: safetensors 0.1.0. Retrieved from https://blogs.rstudio.com/tensorflow/posts/2023-06-15-safetensors/

BibTeX quotation

@misc{safetensors,
  writer = {Falbel, Daniel},
  title = {Posit AI Weblog: safetensors 0.1.0},
  url = {https://blogs.rstudio.com/tensorflow/posts/2023-06-15-safetensors/},
  yr = {2023}
}

Buy now