Skip to content

Releases: tracel-ai/burn

Release list

v0.21.0

v0.21.0 Latest
Latest

Choose a tag to compare

View all tags
@laggui laggui released this 07 May 21:17
v0.21.0
546cacb
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

Summary

Burn 0.21.0 brings 4 months of improvements that make the framework significantly faster and more reliable across the board. The gains span distributed workflows for training large models all the way down to small-model inference, where the reduced framework overhead becomes especially noticeable.

We rethought our distributed computing stack around differentiable collective operations. Kernel selection is now more reliable thanks to better autotuning and a new validation layer, and a project-level burn.toml file lets you tweak those internals (and many others) without recompiling. A reworked device handle reduces framework overhead, and a new burn-dispatch crate simplifies backend selection while paving the way for faster compile times. The release also ships burn-flex, a lightweight eager CPU backend for WebAssembly and embedded targets that replaces burn-ndarray. Finally, we added early off-policy reinforcement learning support and a fresh round of kernel work on GEMV, top-k, and FFT.

For more details, check out the release post on our website.

Changelog

Breaking

We've introduced a couple of breaking changes with this release. The affected areas are detailed in the sections below.

burn-dataset cache directory

To respect platform conventions, we switched from using a hardcoded ~/.cache directory root for downloaded artifacts.

Platform Path
Linux $XDG_CACHE_HOME or ~/.cache
macOS ~/Library/Caches
Windows {FOLDERPATH_LOCAL_APPDATA}

For Linux users without $XDG_CACHE_HOME configured, this change has no effect. The cache directory is still ~/.cache.

Interface Changes

TensorData::shape now stores a Shape instead of a Vec<usize>. Existing binary records using BinFileRecorder or BinBytesRecorder are no not forward-compatible and must be converted before upgrading.

static STATE_ENCODED: &[u8] = include_bytes!("model.bin");

let model: Model<B> = Model::new(&Default::default());

// Old format can still be loaded before upgrade, but must be re-saved in a forward-compatible format.
let record = BinBytesRecorder::<FullPrecisionSettings, &'static [u8]>::default()
    .load(STATE_ENCODED, &Default::default())
    .expect("Failed to decode state");
let model = model.load_record(record);

model.save_file("model.mpk", &NamedMpkFileRecorder::<FullPrecisionSettings>::new()).unwrap();

The module derive macro has been improved, and the Ignored<T> wrapper is now deprecated. For fields that should not considered modules, use #[module(skip)] instead.

pub struct Conv1d<B: Backend> {
-    pub padding: Ignored<PaddingConfig1d>,
+    #[module(skip)]
+    pub padding: PaddingConfig1d,
}

We added support for explicit asymmetric padding. If you were using explicit padding, you must now specify the same value for all pairs. Note that PaddingConfig3d does not support asymmetric padding yet.

// Symmetric (left, right)
- PaddingConfig1d::Explicit(1)
+ PaddingConfig1d::Explicit(1, 1)
// Symmetric (top, left, bottom, right)
- PaddingConfig2d::Explicit(1, 1)
+ PaddingConfig2d::Explicit(1, 1, 1, 1)

The Gelu activation module can now be configured with tanh approximation. This only affects code that instantiated Gelu directly.

- let activation = Gelu;
+ let activation = Gelu::new(); // or Gelu::default()

The position-wise feed-forward module now has a configurable activation function. To keep it backwards compatible with previously saved records, the field is marked as #[module(skip)].

#[derive(Module, Debug)]
pub struct PositionWiseFeedForward<B: Backend> {
    // ...
-   /// GELU activation function.
-   pub gelu: Gelu,
+   /// Activation function.
+   #[module(skip)]
+   pub activation: Activation<B>,
}

The Shape fields are now private and some methods have been renamed. ShapeError has been renamed to MetadataError.

- let b = tensor.shape().dims[0];
+ let b = tensor.shape()[0]

- if let Err(ShapeError::RankMismatch{...}) = lhs.broadcast(&rhs) {
+ if let Err(MetadataError::RankMismatch{...}) = lhs.broadcast(&rhs) {

- let shape = shape.swap(1, 2).unwrap();
+ let shape = shape.swapped(1, 2).unwrap();

- let shape = shape.permute(&[0, 2, 1, 3]).unwrap();
+ let shape = shape.permuted(&[0, 2, 1, 3]).unwrap();

The boolean data type was expanded to include its storage type.

match bool_tensor.dtype() {
-   DType::Bool => todo!(),
+   DType::Bool(BoolStore::Native) => todo!(),
+   DType::Bool(BoolStore::U8) => todo!(),
+   DType::Bool(BoolStore::U32) => todo!(),
    _ => unreachable!(),
}

powf is no longer supported for Int tensors, as it previously relied on incorrect implicit truncation. These operations are now only available for Float tensors.

- let tensor_i = tensor_int.powf(tensor_float);
+ let tensor_f = tensor_int.float().powf(tensor_float);

- let tensor_i = tensor_int.powf_scalar(scalar_float);
+ let tensor_f = tensor_int.float().powf_scalar(scalar_float);

Backend tensor creation and conversion ops now take an explicit output dtype. This removes backend-specific dtype inference and ensures consistent behavior across backends. (Backend implementors only.)

impl BoolTensorOps<Self> for MyBackend {
-    fn bool_empty(shape: Shape, device: &Device<Self>) -> BoolTensor<Self> {
+    fn bool_empty(shape: Shape, device: &Device<Self>, dtype: BoolDType) -> BoolTensor<Self> {
        // use `dtype` instead of inferring internally
    }
-    fn bool_into_int(tensor: BoolTensor<Self>) -> IntTensor<Self> {
+    fn bool_into_int(tensor: BoolTensor<Self>, out_dtype: IntDType) -> IntTensor<Self> {
        // use `dtype` instead of inferring internally
    }
}

Associated types were moved from Backend to BackendTypes. Prefer the type aliases (Device<B>, FloatTensor<B>, etc.) to avoid type resolution issues.

impl BoolTensorOps<Self> for MyBackend {
-    fn bool_empty(shape: Shape, device: &<Self as Backend>::Device, dtype: BoolDType)) -> <Self as Backend>::BoolTensorPrimitive {
+    fn bool_empty(shape: Shape, device: &Device<Self>, dtype: BoolDType) -> BoolTensor<Self> {
    }
}

Module & Tensor

Datasets & Training

Read more

Contributors

jnamika, ovr, and 38 other contributors
Assets 2
Loading

v0.21.0-pre.5

v0.21.0-pre.5 Pre-release
Pre-release

Choose a tag to compare

View all tags
@laggui laggui released this 05 May 20:21
v0.21.0-pre.5
7728590
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

What's Changed

Full Changelog: v0.21.0-pre.4...v0.21.0-pre.5

Contributors

antimora, laggui, and 9 other contributors
Loading

v0.21.0-pre.4

v0.21.0-pre.4 Pre-release
Pre-release

Choose a tag to compare

View all tags
@laggui laggui released this 27 Apr 20:49
v0.21.0-pre.4
8e485cb
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

What's Changed

Full Changelog: v0.21.0-pre.3...v0.21.0-pre.4

Contributors

antimora, ArthurBrussee, and 11 other contributors
Loading

v0.21.0-pre.3

v0.21.0-pre.3 Pre-release
Pre-release

Choose a tag to compare

View all tags
@laggui laggui released this 08 Apr 20:11
v0.21.0-pre.3
52998ac

What's Changed

Full Changelog: v0.21.0-pre.2...v0.21.0-pre.3

Contributors

ovr, antimora, and 16 other contributors
Loading

v0.21.0-pre.2

v0.21.0-pre.2 Pre-release
Pre-release

Choose a tag to compare

View all tags
@nathanielsimard nathanielsimard released this 02 Mar 18:44
v0.21.0-pre.2
ab9f793

What's Changed

Full Changelog: v0.21.0-pre.1...v0.21.0-pre.2

Contributors

antimora, wingertge, and 15 other contributors
Loading

v0.21.0-pre.1

v0.21.0-pre.1 Pre-release
Pre-release

Choose a tag to compare

View all tags
@nathanielsimard nathanielsimard released this 09 Feb 22:13
v0.21.0-pre.1
3fa8dfa
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

What's Changed

Contributors

jnamika, antimora, and 20 other contributors
Loading

v0.20.1

v0.20.1

Choose a tag to compare

View all tags
@laggui laggui released this 23 Jan 17:43
v0.20.1
75b7881

Bug Fixes & Improvement

Contributors

laggui
Loading

v0.20.0

v0.20.0

Choose a tag to compare

View all tags
@laggui laggui released this 15 Jan 16:08
v0.20.0
3475ba8

Summary

This release marks a major turning point for the ecosystem with the introduction of CubeK. Our goal was to solve a classic challenge in deep learning: achieving peak performance on diverse hardware without maintaining fragmented codebases.

By unifying CPU and GPU kernels through CubeCL, we've managed to squeeze maximum efficiency out of everything from NVIDIA Blackwell GPUs to standard consumer CPUs.

Beyond performance, this release makes the library more robust, flexible, and significantly easier to debug.

This release also features a complete overhaul of the ONNX import system, providing broader support for a wide range of ONNX models. In addition, various bug fixes and new tensor operations enhance stability and usability.

For more details, check out the release post on our website.

Changelog

Breaking

We've introduced a couple of breaking API changes with this release. The affected interfaces are detailed in the sections below.

Training

We refactored burn-train to better support different abstractions and custom training strategies. As part of this,
the LearnerBuilder has been replaced by the LearningParadigm flow:

- let learner = LearnerBuilder::new(ARTIFACT_DIR)
+ let training = SupervisedTraining::new(ARTIFACT_DIR, dataloader_train, dataloader_valid)
        .metrics((AccuracyMetric::new(), LossMetric::new()))
        .num_epochs(config.num_epochs)
-       .learning_strategy(burn::train::LearningStrategy::SingleDevice(device))
-       .build(model, config.optimizer.init(), lr_scheduler.init().unwrap());
+       .summary();
 
- let result = learner.fit(dataloader_train, dataloader_valid);
+ let result = training.launch(Learner::new(
+      model,
+      config.optimizer.init(),
+      lr_scheduler.init().unwrap(),
+ ));

Interface Changes

The scatter and select_assign operations now require an IndexingUpdateOp to specify the update behavior.

- let output = tensor.scatter(0, indices, values);
+ let output = tensor.scatter(0, indices, values, IndexingUpdateOp::Add);

API calls for slice, slice_assign, and slice_fill no longer require const generics for dimensions, which cleans up the syntax quite a bit:

- let prev_slice = tensor.slice::<[Range<usize>; D]>(slices.try_into().unwrap());
+ let prev_slice = tensor.slice(slices.as_slice());

The grid_sample_2d operation now supports different options.
To preserve the previous behavior, make sure to specify the matching options:

- let output = tensor.grid_sample_2d(grid, InterpolateMode::Bilinear);
+ let options = GridSampleOptions::new(InterpolateMode::Bilinear)
+     .with_padding_mode(GridSamplePaddingMode::Border)
+     .with_align_corners(true);
+ let output = tensor.grid_sample_2d(grid, options);

The QuantStore variants used in QuantScheme have been updated to support a packing dimension.

  pub enum QuantStore {
      /// Native quantization doesn't require packing and unpacking.
      Native,
+     /// Store packed quantized values in a natively supported packing format (i.e. e2m1x2).
+     PackedNative(usize),
      /// Store packed quantized values in a 4-byte unsigned integer.
-     U32,
+     PackedU32(usize),
 }

Finally, Shape no longer implements IntoIterator. If you need to iterate by-value over dimensions, access the dims field directly.

- for s in shape {
+ for s in shape.dims {

Module & Tensor

Datasets & Training

Backends

Bug Fixes

Documentation & Examples

Fixes

ONNX Support

Enhancements

Refactoring

  • chore: Update to batch caching PR for cubecl (#3948) @wingertge
  • Refactor IR to define outputs as a function of the operation (#3877) ...
Read more

Contributors

khoek, antimora, and 19 other contributors
Loading

v0.20.0-pre.6

v0.20.0-pre.6 Pre-release
Pre-release

Choose a tag to compare

View all tags
@nathanielsimard nathanielsimard released this 18 Dec 21:27
v0.20.0-pre.6
91dd62c
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

What's Changed

Contributors

antimora, wingertge, and 9 other contributors
Loading

v0.20.0-pre.5

v0.20.0-pre.5 Pre-release
Pre-release

Choose a tag to compare

View all tags
@nathanielsimard nathanielsimard released this 08 Dec 14:53
v0.20.0-pre.5
42edc63
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

What's Changed

Contributors

antimora, wingertge, and 6 other contributors
Loading