FastGelu float16 #621
Merged
FastGelu float16 #621
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
wenbingl
approved these changes
Dec 11, 2023
| #include "onnxruntime_c_api.h" | ||
| #if ORT_API_VERSION >= 16 | ||
|
|
||
| #include "onnxruntime_float16.h" |
There was a problem hiding this comment.
is this file shipping with ort C++ package?
There was a problem hiding this comment.
yes it is, since 1.16.
| const T* bias_data = bias.has_value() ? (*bias)->Data() : nullptr; | ||
| auto bias_length = bias.has_value() ? (*bias)->NumberOfElement() : 0; | ||
| using TT = typename CudaT<T>::MappedType; | ||
| LaunchFastGeluKernel<TT>(reinterpret_cast<cudaStream_t>(ctx.cuda_stream), |
There was a problem hiding this comment.
should the return error code be handled here?
There was a problem hiding this comment.
Good point - will report it in coming iteration.
| ] | ||
|
|
||
| input0 = helper.make_tensor_value_info( | ||
| 'x', onnx_proto.TensorProto.FLOAT16, []) |
There was a problem hiding this comment.
dumb question, which fp16 was tested here, MFloat16 or BFloat16?
There was a problem hiding this comment.
It is MFloat16.
For BFloat16, we need to test it by native cases since the type is not exposed via python.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Add float16 support for contrib cuda ops.