Commit graph

12 commits

Author SHA1 Message Date
Guenther Schmuelling
55a6986d38
optimize skiplayernorm (#20551)
SkipSimplifiedLayerNormalization used in phi3 comes down from 222usec to
14usec
2024-05-08 08:40:03 -07:00
Guenther Schmuelling
497a627a69
fix fp16 for skiplayernorm (#20381) 2024-04-19 12:12:02 -07:00
Yulong Wang
50bd4571ac
[js/web] support SimplifiedLayerNorm and SkipSimplifiedLayerNorm (#20277)
### Description
Support operator `SimplifiedLayerNorm` and `SkipSimplifiedLayerNorm` for
WebGPU backend.
2024-04-11 14:08:50 -07:00
Xu Xing
d73131cf0f
[js/webgpu] Use DataType as uniform cpu type (#19281)
This saves turning data type to string by tensorDataTypeEnumToString.
2024-01-30 21:05:08 -08:00
satyajandhyala
a33b5bd1fa
[JS/WebGPU] Added Uniforms to SkipLayerNorm. (#18788)
### Description
Added Uniforms to SkipLayerNorm



### Motivation and Context
<!-- - Why is this change required? What problem does it solve?
- If it fixes an open issue, please link to the issue here. -->
Improve performance

---------

Co-authored-by: Yulong Wang <7679871+fs-eire@users.noreply.github.com>
2024-01-25 01:12:21 +05:30
Jiajie Hu
acba63c36a
[js/webgpu] Change A/sqrt(B) to A*inverseSqrt(B) in normalization ops (#19101)
### Description
Change `A / sqrt(B)` to `A * inverseSqrt(B)` in BatchNormalization,
InstanceNormalization, LayerNormalization and SkipLayerNormalization.

### Motivation and Context
For the same reason as the existence of the `inverseSqrt` built-in in
WebGPU spec.
2024-01-12 00:08:16 -08:00
Arthur Islamov
22947109f2
[js/web] FP16 LayerNorm, InstanceNorm, SkipLayerNorm (#17630)
### Description
This PR includes fixes for Norm operations to support FP16 and also some
optimizations to use vec2/vec4 if possible
2023-10-18 10:47:41 -07:00
Yulong Wang
d532645bed
[js/webgpu] revise uniform support (#17871)
### Description
<!-- Describe your changes. -->

work for items (2) and (3) in #17860
2023-10-11 16:41:46 -07:00
Yulong Wang
d9b9c5a537
[js/webgpu] support using uniform buffer (#17803)
### Description
support using uniform buffer.

This PR allows to use uniform buffer in shader program, so that some
runtime information (eg. input/output shape) is no longer need to be
hardcoded into shader code.

There are 2 commits in this PR:
-
[667f31c](667f31c83d):
framework changes to support uniform buffer, as well as updates in
program manager, gpu data manager and indices helper.
-
[09e1d2a](09e1d2ad1d):
an example change for operator `Transpose` to use input's rank-only
instead of dims as shader key. With this change, model mobilenetv2-12
shader compile times dropped from 71 to 52.
2023-10-10 00:31:12 -07:00
Yulong Wang
9aafbe3feb
[js/web] revise TensorView (#17473)
### Description

This change:
- removes the unused `Tensor` types declared in
/js/web/lib/wasm/jsep/tensor.ts
- removes duplicated util functions in  /js/web/lib/wasm/jsep/tensor.ts
- renames /js/web/lib/wasm/jsep/**tensor.ts** to
/js/web/lib/wasm/jsep/**tensor-view.ts** and update corresponding
references. It was kind of confusing that we have multiple `Tensor`
types defined in different places also we have multiple `tensor.ts`
source files.

This is one of the prerequisites for supporting IO binding for WebGPU
buffer in onnxruntime-web.

list of prerequisites PRs:
https://github.com/microsoft/onnxruntime/pull/17465
https://github.com/microsoft/onnxruntime/pull/17469
https://github.com/microsoft/onnxruntime/pull/17470
https://github.com/microsoft/onnxruntime/pull/17472
https://github.com/microsoft/onnxruntime/pull/17473 (this one)
2023-09-14 21:14:44 -07:00
Yulong Wang
fb51faea64
[js/webgpu] fix 2 build breaks introduced in merge (#17273)
### Description
fix 2 build breaks introduced in merge. Fixes web build
2023-08-23 18:09:50 -07:00
satyajandhyala
7d1a5635a0
[JS/Web] Added SkipLayerNormalization operator. (#17102)
### Description
Add SkipLayerNormalization operator to JSEP.



### Motivation and Context
<!-- - Why is this change required? What problem does it solve?
- If it fixes an open issue, please link to the issue here. -->
2023-08-18 09:59:03 -07:00