saymrwulf/onnxruntime
1
0
Fork 0
mirror of https://github.com/saymrwulf/onnxruntime.git synced 2026-07-03 03:58:54 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Fix MS domain QuantizeLinear and DequantizeLinear type registrations … (#16298)

Browse source 
This fixes the type lists used to register DML kernels for Microsoft
domain QuantizeLinear and DequantizeLinear. These previously did not
include FP16 and incorrectly used the same type list for both operators.

The new type lists are the same as opset 19 ONNX which aren't
implemented yet in the DML EP.
This commit is contained in:
Jeff Bloomfield 2023-06-15 18:21:56 -07:00 committed by GitHub
parent 188d5f5398
commit 6949cfaf94
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
2 changed files with 6 additions and 5 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

4
docs/OperatorKernels.md
View file

@ -1223,7 +1223,7 @@ Do not modify directly.*
|BiasGelu|*in* A:**T**<br> *in* B:**T**<br> *out* C:**T**|1+|**T** = tensor(float), tensor(float16)|
|BiasSplitGelu|*in* X:**T**<br> *in* bias:**T**<br> *out* Y:**T**|1+|**T** = tensor(float), tensor(float16)|
|ConvTransposeWithDynamicPads|*in* X:**T**<br> *in* W:**T**<br> *in* Pads:**tensor(int64)**<br> *in* B:**T**<br> *out* Y:**T**|1+|**T** = tensor(float), tensor(float16)|
|DequantizeLinear|*in* x:**T1**<br> *in* x_scale:**T2**<br> *in* x_zero_point:**T1**<br> *out* y:**T2**|1+|**T1** = tensor(float)<br/> **T2** = tensor(uint8)|
|DequantizeLinear|*in* x:**T1**<br> *in* x_scale:**T2**<br> *in* x_zero_point:**T1**<br> *out* y:**T2**|1+|**T1** = tensor(int32), tensor(int8), tensor(uint8)<br/> **T2** = tensor(float), tensor(float16)|
|EmbedLayerNormalization|*in* input_ids:**T1**<br> *in* segment_ids:**T1**<br> *in* word_embedding:**T**<br> *in* position_embedding:**T**<br> *in* segment_embedding:**T**<br> *in* gamma:**T**<br> *in* beta:**T**<br> *in* mask:**T1**<br> *in* position_ids:**T1**<br> *out* output:**T**<br> *out* mask_index:**T1**<br> *out* embedding_sum:**T**|1+|**T** = tensor(float), tensor(float16)|
|FusedMatMul|*in* A:**T**<br> *in* B:**T**<br> *out* Y:**T**|1+|**T** = tensor(float), tensor(float16)|
|FusedMatMulActivation|*in* A:**T**<br> *in* B:**T**<br> *out* Y:**T**|1+|**T** = tensor(float), tensor(float16)|
@ -1233,7 +1233,7 @@ Do not modify directly.*
|NhwcConv|*in* X:**T**<br> *in* W:**T**<br> *in* B:**T**<br> *out* Y:**T**|1+|**T** = tensor(float), tensor(float16)|
|QLinearAdd|*in* A:**T**<br> *in* A_scale:**tensor(float)**<br> *in* A_zero_point:**T**<br> *in* B:**T**<br> *in* B_scale:**tensor(float)**<br> *in* B_zero_point:**T**<br> *in* C_scale:**tensor(float)**<br> *in* C_zero_point:**T**<br> *out* C:**T**|1+|**T** = tensor(int8), tensor(uint8)|
|QLinearSigmoid|*in* X:**T**<br> *in* X_scale:**tensor(float)**<br> *in* X_zero_point:**T**<br> *in* Y_scale:**tensor(float)**<br> *in* Y_zero_point:**T**<br> *out* Y:**T**|1+|**T** = tensor(int8), tensor(uint8)|
|QuantizeLinear|*in* x:**T1**<br> *in* y_scale:**T1**<br> *in* y_zero_point:**T2**<br> *out* y:**T2**|1+|**T1** = tensor(float)<br/> **T2** = tensor(uint8)|
|QuantizeLinear|*in* x:**T1**<br> *in* y_scale:**T1**<br> *in* y_zero_point:**T2**<br> *out* y:**T2**|1+|**T1** = tensor(float), tensor(float16), tensor(int32)<br/> **T2** = tensor(int8), tensor(uint8)|
|QuickGelu|*in* X:**T**<br> *out* Y:**T**|1+|**T** = tensor(float), tensor(float16)|
|SkipLayerNormalization|*in* input:**T**<br> *in* skip:**T**<br> *in* gamma:**T**<br> *in* beta:**T**<br> *in* bias:**T**<br> *out* output:**T**<br> *out* mean:**U**<br> *out* inv_std_var:**U**<br> *out* input_skip_bias_sum:**T**|1+|**T** = tensor(float), tensor(float16)|
| |

7
onnxruntime/core/providers/dml/DmlExecutionProvider/src/Operators/OperatorRegistration.cpp
View file

@ -510,8 +510,9 @@ constexpr static std::array<SupportedTensorDataTypes, 2> supportedTypeListScatte
constexpr static std::array<SupportedTensorDataTypes, 1> supportedTypeListScatterGatherND = { SupportedTensorDataTypes::AllScalars };
constexpr static std::array<SupportedTensorDataTypes, 2> supportedTypeListSlice10 = { SupportedTensorDataTypes::AllScalars, SupportedTensorDataTypes::Int32 | SupportedTensorDataTypes::Int64 };
constexpr static std::array<SupportedTensorDataTypes, 2> supportedTypeListQuantizeLinear = { SupportedTensorDataTypes::Float32 | SupportedTensorDataTypes::Int32, SupportedTensorDataTypes::UInt8 | SupportedTensorDataTypes::Int8 };
constexpr static std::array<SupportedTensorDataTypes, 2> supportedTypeListQuantizeLinear19 = { SupportedTensorDataTypes::Float16to32 | SupportedTensorDataTypes::Int32, SupportedTensorDataTypes::UInt8 | SupportedTensorDataTypes::Int8 };
constexpr static std::array<SupportedTensorDataTypes, 1> supportedTypeListDequantizeLinear = { SupportedTensorDataTypes::UInt8 | SupportedTensorDataTypes::Int8 | SupportedTensorDataTypes::Int32 };
constexpr static std::array<SupportedTensorDataTypes, 2> supportedTypeListQuantize = { SupportedTensorDataTypes::Float32, SupportedTensorDataTypes::UInt8 };
constexpr static std::array<SupportedTensorDataTypes, 2> supportedTypeListDequantizeLinear19 = { SupportedTensorDataTypes::UInt8 | SupportedTensorDataTypes::Int8 | SupportedTensorDataTypes::Int32, SupportedTensorDataTypes::Float16to32 };
constexpr static std::array<SupportedTensorDataTypes, 2> supportedTypeListIsNan = { SupportedTensorDataTypes::Float16to32, SupportedTensorDataTypes::Bool };
constexpr static std::array<SupportedTensorDataTypes, 2> supportedTypeListIsInf = { SupportedTensorDataTypes::Float32, SupportedTensorDataTypes::Bool };
constexpr static std::array<SupportedTensorDataTypes, 2> supportedTypeListConstantOfShape = { SupportedTensorDataTypes::Int64, SupportedTensorDataTypes::AllScalars };
@ -762,8 +763,8 @@ constexpr static OperatorRegistrationInformation operatorRegistrationInformation
{REG_INFO( 13, QuantizeLinear, typeNameListTwo, supportedTypeListQuantizeLinear, DmlGraphSupport::Supported)},
{REG_INFO( 10, DequantizeLinear, typeNameListDefault, supportedTypeListDequantizeLinear, DmlGraphSupport::Supported)},
{REG_INFO( 13, DequantizeLinear, typeNameListDefault, supportedTypeListDequantizeLinear, DmlGraphSupport::Supported)},
{REG_INFO_MS( 1, QuantizeLinear, typeNameListTwo, supportedTypeListQuantize, DmlGraphSupport::Supported)},
{REG_INFO_MS( 1, DequantizeLinear, typeNameListTwo, supportedTypeListQuantize, DmlGraphSupport::Supported)},
{REG_INFO_MS( 1, QuantizeLinear, typeNameListTwo, supportedTypeListQuantizeLinear19, DmlGraphSupport::Supported)},
{REG_INFO_MS( 1, DequantizeLinear, typeNameListTwo, supportedTypeListDequantizeLinear19, DmlGraphSupport::Supported)},
{REG_INFO( 9, Sign, typeNameListDefault, supportedTypeListFloat16to32Ints8to64, DmlGraphSupport::Supported)},
{REG_INFO( 13, Sign, typeNameListDefault, supportedTypeListFloat16to32Ints8to64, DmlGraphSupport::Supported)},
{REG_INFO( 9, IsNaN, typeNameListTwo, supportedTypeListIsNan, DmlGraphSupport::Supported)},