dnrops
/
burn
mirror of https://github.com/tracel-ai/burn.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
873 Commits 39 Branches 19 Tags 101 MiB
Commit Graph

330 Commits

Author SHA1 Message Date
yurzhang 419e53bc42
Support signed value for `Tensor::arange` (#1238) 2024-02-07 09:33:01 -05:00
Joshua Ferguson 1d84eb775e
`NaN` and `Inf` detection in `assert_approx_eq` (#1209) 2024-02-02 14:53:34 -05:00
Nathaniel Simard 9df20712bf
Fix docs (#1229) 2024-02-01 14:50:38 -05:00
Sylvain Benner c4e512d176
Create doc feature in all burn crates and use them for docs.rs build (#1212) 2024-02-01 12:47:46 -05:00
Sylvain Benner 4aa13d6b25
Bump Burn version to 0.13 (#1211) 2024-01-31 16:01:20 -05:00
Louis Fortier-Dubois e03facc5de
fix powf bugs (#1207) 2024-01-31 11:38:45 -05:00
Joshua Ferguson 07c69bd29a
Powf fix2 (#1193) 2024-01-31 10:19:15 -05:00
Joshua Ferguson 4a70a0f8bc
renaming FloatTensor Ops, Primitives, and maybe functions (#1174) 2024-01-27 10:04:50 -05:00
Joshua Ferguson 3b7d9feede
Elementwise pow op (#1133) 2024-01-24 09:46:57 -05:00
James MacGlashan a752cc172c
Allow raw tensors to be serialized/deserialized directly with serde (#1041) 2024-01-20 11:25:49 -05:00
wcshds a5bdf38c92
fix the problem of sigmoid gradient generating NaN (#1140) 
* use sigmoid derivative formulas

* add test

* fix test error

* move sigmoid to tensor/ops/activation.rs

* use full precision in the default implementation

* rename the param of `sigmoid_backward`
 2024-01-16 16:20:18 -05:00
Nathaniel Simard 0a85744d9b
Refactor/fusion stream (#1135) 2024-01-13 14:17:43 -05:00
Mykolas Peteraitis e132e21816
Small fix: documentation of Tensor::stack (#1105) 2024-01-08 08:38:27 -05:00
Nathaniel Simard f0e10f1499
Refactor/fusion optimization (#1104) 2024-01-06 14:22:50 -05:00
Kirill Mavreshko 97297538b1
Remove _devauto fuctions (#518) (#1110) 2024-01-06 13:36:34 -05:00
unrenormalizable 40ec289a92
Making Embedding.weight public (#1094) 2023-12-22 18:11:38 -05:00
Nathaniel Simard d82e6b157b
Fix tests (#1089) 
* Fix tests

* Fix fmt

* Fix CI
 2023-12-21 13:06:19 -05:00
Kirill Mavreshko 1fd07fcb4a
Explicit device tensors (#1081) 2023-12-20 17:49:59 -05:00
Kelvin Wu 7c6f017c98
Implement chunk for different backends (#1032) 2023-12-20 13:35:59 -05:00
pacowong c1cb77ac2e
Add mish and softplus activation function (#1071) 2023-12-20 12:41:23 -05:00
Nathaniel Simard 75062c51e0
Refactor/wgpu/binary (#1078) 
* Refactor binary

* Fix

* Oups

* Preparation for cmp

* Refactor comparison

* Remove templates

* Cleanup

* Cleanup

* Fix typo

* Code review
 2023-12-19 14:44:55 -05:00
AuruTus fc97a28f16
fix/ndarray: remove reversed axes check (#1058) 2023-12-14 18:09:24 -05:00
Louis Fortier-Dubois 4608cd97d2
Feat/wgpu/repeat (#1068) 2023-12-14 18:08:37 -05:00
Nathaniel Simard 2ca1fd5a92
Fix bugs (#1064) 2023-12-12 16:19:01 -05:00
Alex Errant 610d64095e
cargo +nightly fmt (#1017) 2023-12-12 13:29:06 -05:00
David Chavez 71d3c1d142
chore(infra): Share some properties across workspace (#1039) 2023-12-12 09:39:07 -05:00
Louis Fortier-Dubois 8fc52113bc
Chore/bump v12 (#1048) 2023-12-04 10:47:54 -05:00
Louis Fortier-Dubois 3088c466a5
patch 0.11.1 (#1047) 2023-12-04 10:18:30 -05:00
Louis Fortier-Dubois 4d7944912d
fix gatter typo (#1036) 2023-12-01 15:28:25 -05:00
Nathaniel Simard ab1b5890f5
Chore/release (#1031) 2023-12-01 14:33:28 -05:00
Nathaniel Simard 670280dda2
Feat/fusion/cache (#1020) 2023-12-01 12:05:11 -05:00
Nathaniel Simard b0de56da29
Fix double broadcast with tch (#1026) 
* Fix double broadcast with tch

* More fixes

* Fix clippy warm
 2023-12-01 10:02:57 -05:00
wcshds 1f18cf4adf
implement DoubleEndedIterator trait for DimIter (#1022) 
* implement DoubleEndedIterator for DimIter

* return early in next() and next_back()

* return early in next() and next_back()

* remove #[rustfmt::skip]
 2023-11-30 12:26:16 -06:00
Will Brickner 03af140e12
Implement Quiet Softmax (`Attention Is Off By One`) (#692) 
* Added quiet_softmax

* Undid bad formatting

---------

Co-authored-by: Dilshod Tadjibaev <939125+antimora@users.noreply.github.com>
 2023-11-30 12:58:30 -05:00
David Chavez fda9e5d388
feature(tensor): Add triu and tril ops (#1002) 2023-11-29 09:39:42 -05:00
Louis Fortier-Dubois aa3180d0c7
fix/docs/chunk (#1006) 2023-11-29 09:38:43 -05:00
David Chavez 929b1786bb
feature(tensor): Add chunk op (#998) 2023-11-27 09:58:43 -05:00
David Chavez 2fdf9a3591
feature(tensor): Add Tensor::stack (#1004) 2023-11-27 09:57:31 -05:00
David Chavez 2928657622
chore(tensors): fix narrow error messages (#997) 2023-11-24 15:13:07 -05:00
Zsombor e9f4680153
Fix handling of the dimension parameter in max_dim/min_dim (#995) 
And add a couple of tests
 2023-11-24 13:04:29 -05:00
David Chavez f09baadfc5
feature(tensor): Add narrow op (#996) 
* Add narrow methods

* Revert "Add narrow methods"

This reverts commit 9371d87c79.

* Implement a shared version of narrow

* Correct test case

* Update book

* Improve tests
 2023-11-24 08:24:06 -05:00
Nathaniel Simard 3d6c738776
Refactor/fusion/graph (#988) 2023-11-22 09:55:42 -05:00
Nathaniel Simard cabbaab0c4
Fix/constant tensors (#984) 
* Generalize autodiff tensor

* Can have int const module

* Update example

* Support no-std with burn-import

* Fix typos

* Fix alloc problems

* Revert burn-import changes

* Fix examples

* Support Int and Bool Params

* Fix

* Add comment
 2023-11-21 15:27:28 -06:00
David Chavez 49e16b6834
feature(tensor): Add unsqueeze_dim helper (#966) 2023-11-20 10:29:40 -05:00
Jean-Gab e5c6044062
doc(burn-tensor): Add examples to slice operation to help noobs like me understand what it does (#880) 
* doc(burn-tensor): Add examples to slice operation to help noobs like me understand what it does

* doc(burn-tensor): Improve slice documentation with more comments and clearer examples

* Fix formatting

* Fix typo.

---------

Co-authored-by: jeangab <jg@nationtech.io>
Co-authored-by: Dilshod Tadjibaev <939125+antimora@users.noreply.github.com>
 2023-11-19 23:13:56 -06:00
Luni-4 ec9df53d4c
ci/Fix `cargo clippy` action (#942) 2023-11-16 19:35:38 -05:00
Nathaniel Simard 24014aca33
WGPU: Support elemwise operation fusion (#948) 2023-11-15 15:13:37 -05:00
Zsombor 4fc0c27e31
Implement tensor.recip() function to calculate elementwise reciprocals (#953) 2023-11-15 09:17:32 -05:00
chenkun 2614944afa
fix approximately equal precision issue in test code (#954) 2023-11-13 15:35:24 -05:00
Nathaniel Simard 322480b744
Feat/op fusion decorator (#939) 
* WIP

* Impl backend decorator

* WIP

* WIP

* WIP

* WIP

* WIP

* WIP

* Refactor

* Handle graph single ops execution

* WIP

* Starting to get concrete

* WIP

* Fix locator

* Implement add ops

* Start implementing ops

* Add more ops

* Add more ops

* More float ops

* Almost finish float ops

* Almost done with Int

* Some fix

* Into float

* Implement bool ops

* Almost done with MVP

* Fix adaptive pooling

* Add fusion as backend

* Fix memory leak

* Fix

* WIP Doc

* Doc all ops enum

* Initial docs

* Clippy

* Clippy v2

* Fix typos

* Fix doc

* Fix feature flags

* Add missing ops

* Some cleanup

* Revert u128 id

* cosmetic fixes

---------

Co-authored-by: louisfd <louisfd94@gmail.com>
 2023-11-09 21:21:41 -05:00