saymrwulf/onnxruntime
1
0
Fork 0
mirror of https://github.com/saymrwulf/onnxruntime.git synced 2026-07-23 19:32:23 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions
onnxruntime/js/web/lib/wasm/jsep/webgpu/ops
History
Jiajia Qin ccbe264a39
[js/webgpu] Add LeakyRelu activation for fusedConv (#19369) 
### Description
This PR 1) adds LeakyRelu activation for fusedConv; 2) makes `vec4<f16>`
value work with `float32` uniforms attributes.

For example:
`clamp(value, vec4<f16>(uniforms.clip_min),
vec4<f16>(uniforms.clip_max)` will throw compilation errors since
`uniforms.clip_min` and `uniforms.clip_min` are `f32` not `f16`. So we
need to change it to `clamp(value, vec4<f16>(f16(uniforms.clip_min)),
vec4<f16>(f16(uniforms.clip_max))`

And above problem was introduced when we make activation attributes as
uniforms instead of constant.

BTW, after adding LeakyRelu, `realesrgan-t256` model can pass.
2024-02-02 09:06:38 -08:00
..
3rd-party [js/webgpu] Add LeakyRelu activation for fusedConv (#19369) 2024-02-02 09:06:38 -08:00
argminmax.ts [JS/Web] Added uniforms to Reduce, Resize and Split Ops. (#18727) 2023-12-12 11:12:23 -08:00
attention.ts [js/webgpu] Use DataType as uniform cpu type (#19281) 2024-01-30 21:05:08 -08:00
batch-norm.ts [js/webgpu] Use DataType as uniform cpu type (#19281) 2024-01-30 21:05:08 -08:00
bias-add.ts
bias-split-gelu.ts [JS/Web] Resize & BiasSplitGelu fp16 support (#18536) 2023-11-22 12:12:07 -08:00
binary-op.ts [js/webgpu] Refactor createTensorShapeVariables (#18883) 2024-02-01 17:59:00 -08:00
common.ts [js/webgpu] Refactor createTensorShapeVariables (#18883) 2024-02-01 17:59:00 -08:00
concat.ts [js/webgpu] Use DataType as uniform cpu type (#19281) 2024-01-30 21:05:08 -08:00
conv-grouped.ts [js/webgpu] Add LeakyRelu activation for fusedConv (#19369) 2024-02-02 09:06:38 -08:00
conv-transpose.ts [js/webgpu] Support uniforms for conv, conv transpose, conv grouped (#18753) 2024-01-25 15:37:05 -08:00
conv.ts [js/webgpu] Support uniforms for conv, conv transpose, conv grouped (#18753) 2024-01-25 15:37:05 -08:00
cumsum.ts [js/webgpu] Refactor createTensorShapeVariables (#18883) 2024-02-01 17:59:00 -08:00
einsum.ts [js/webgpu] Use DataType as uniform cpu type (#19281) 2024-01-30 21:05:08 -08:00
expand.ts [js/webgpu] Refactor createTensorShapeVariables (#18883) 2024-02-01 17:59:00 -08:00
fuse-utils.ts [js/webgpu] Add LeakyRelu activation for fusedConv (#19369) 2024-02-02 09:06:38 -08:00
gather-elements.ts [js/webgpu] Refactor createTensorShapeVariables (#18883) 2024-02-01 17:59:00 -08:00
gather.ts [js/webgpu] Refactor createTensorShapeVariables (#18883) 2024-02-01 17:59:00 -08:00
gemm.ts [js/webgpu] Use DataType as uniform cpu type (#19281) 2024-01-30 21:05:08 -08:00
instance-norm.ts [js/webgpu] Refactor createTensorShapeVariables (#18883) 2024-02-01 17:59:00 -08:00
layer-norm.ts [js/webgpu] Use DataType as uniform cpu type (#19281) 2024-01-30 21:05:08 -08:00
matmul.ts [js/webgpu] Add LeakyRelu activation for fusedConv (#19369) 2024-02-02 09:06:38 -08:00
multi-head-attentiion.ts [js/webgpu] Use DataType as uniform cpu type (#19281) 2024-01-30 21:05:08 -08:00
pad.ts [js/webgpu] Refactor createTensorShapeVariables (#18883) 2024-02-01 17:59:00 -08:00
pool.ts [js/webgpu] Refactor createTensorShapeVariables (#18883) 2024-02-01 17:59:00 -08:00
range.ts [js/webgpu] Use DataType as uniform cpu type (#19281) 2024-01-30 21:05:08 -08:00
reduce-shared.ts [js/webgpu] Use DataType as uniform cpu type (#19281) 2024-01-30 21:05:08 -08:00
reduce.ts [js/webgpu] Refactor createTensorShapeVariables (#18883) 2024-02-01 17:59:00 -08:00
resize.ts [js/webgpu] Refactor createTensorShapeVariables (#18883) 2024-02-01 17:59:00 -08:00
skip-layer-norm.ts [js/webgpu] Use DataType as uniform cpu type (#19281) 2024-01-30 21:05:08 -08:00
slice.ts [js/webgpu] Refactor createTensorShapeVariables (#18883) 2024-02-01 17:59:00 -08:00
softmax.ts [js/webgpu] Use DataType as uniform cpu type (#19281) 2024-01-30 21:05:08 -08:00
split.ts [js/webgpu] Refactor createTensorShapeVariables (#18883) 2024-02-01 17:59:00 -08:00
tile.ts [js/webgpu] Refactor createTensorShapeVariables (#18883) 2024-02-01 17:59:00 -08:00
transpose.ts [js/webgpu] Refactor createTensorShapeVariables (#18883) 2024-02-01 17:59:00 -08:00
unary-op.ts [js/webgpu] Use DataType as uniform cpu type (#19281) 2024-01-30 21:05:08 -08:00
where.ts [js/webgpu] Refactor createTensorShapeVariables (#18883) 2024-02-01 17:59:00 -08:00