### Description
Previously, MultiHeadAttention supports relative position bias of shape
[1, N, S, T] or [B, N, S, T], and DecoderMaskedMultiHeadAttention
supports [1, N, S, T]. This will extend the support to allow [1, N, S,
T], [B, N, S, T], [B, 1, S, T] and [1, 1, S, T] for CUDA and CPU EPs.
- [x] Rename the input of "relative position bias" to "attention bias"
because it can also be used for other types of bias, like ALiBi
(Attention with Linear Biases) or attention mask.
- [x] Update unfused kernel to support broadcasting 2nd dimension of
attention bias.
- [x] Update efficient attention to support broadcasting 2nd dimension
of attention bias.
- [x] Update operators (MultiHeadAttention,
DecoderMaskedMultiHeadAttention, Attention, PackedAttention,
PackedMultiHeadAttention) to support broadcast attention bias on CUDA
and CPU EPs.
- [x] Update ROCm, DML and WebGPU naming to be consistent. (Note that
those EPs do not support broadcasting attention_bias for now).
- [x] Add attention bias tests for MultiHeadAttention.
- [x] Update operator documents
- [x] Update benchmark script
Other changes:
* Fix some checks in multihead-attention.ts
* Add helper functions to dump tensors given dimensions.
### Description
Add a gather that supports block-quantized input data.
### Motivation and Context
To support Web inference scenario with quantized vocabulary embeddings.
### Description
This PR registers the ReduceMin-20 operator to the DML EP.
### 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. -->
Add SparseAttention cpu implementation.
- [x] Refactoring GQAAttentionBase
- [x] Add SparseAttention implementation
- [x] Add test cases
This is unfused version. Flash attention version will be added later.
### Description
<!-- Describe your changes. -->
Trilu<bool> is used by phi-3 when exported with torch.onnx.export.
### 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. -->
### Description
- 4-bit QuantizeLinear(21). **Blocked quantization still missing (i.e.,
do not support the new `block_size` attribute)**
- 4-bit DequantizeLinear(21). **Blocked dequantization still missing
(i.e., do not support the new `block_size` attribute)**
- 4-bit Transpose(21).
- Update quantization tool with int4 types.
- Disable QDQ fusions for 4-bit types. See:
https://github.com/microsoft/onnxruntime/blob/main/onnxruntime/core/optimizer/qdq_transformer/selectors_actions/qdq_selector_action_transformer.cc
- MLAS 4-bit quantization kernels for intel, neon, powerpc.
##### Notes
To calculate a tensor's storage size, we normally get the number of
elements from the shape (i.e., `tensor_shape.Size()`) and multiply by
the size of a single element. This does not directly work for sub-byte
elements like int4 as each element in a `Tensor<Int4x2>` stores **two**
packed int4 elements in a byte. The `Tensor::
CalculateTensorStorageSize` should be called to perform the correct
calculation for any tensor element type.
### Motivation and Context
ONNX 1.16 added the int4 and uint4 types. This initial PR adds the int4
type to ORT and adds int4 implementations for the Quant, Dequant, and
Transpose ops on CPU EP. We still need to add int4 support for many ops
and execution providers. See the ONNX 1.16 release notes:
https://github.com/onnx/onnx/releases.
### Description
Make the operator more flexible:
(1) Decouple the max sequence length of rotary cache, kv cache and block
mask. They are allowed to have different values.
(2) Replace block_mask dense by CSR format (block_row_indices and
block_col_indices) to improve performance.
(3) Mark past_key and past_value as required inputs since we need them
to compute the shape of present_key and present_value.
### Motivation and Context
(1) LongRoPE has short and long rotary cache, which has different
length.
(2) Most users do not have enough GPU memory to run maximum sequence
length 128K. This change allows user to use smaller kv cache length to
test without hitting out of memory.
### Description
<!-- Describe your changes. -->
This PR registers DFT-20 to the DML EP.
### 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. -->
### Description
Add CUDA implementation for block sparse attention for Phi-3-small.
Block sparse attention was proposed in [Sparse
Transformers](https://arxiv.org/pdf/1904.10509) by OpenAI, and also
adopted in [BigBird](https://arxiv.org/pdf/2007.14062) with different
sparse layout.
In Phi-3-small, the sparse layout is static, and works with
unidirectional (causal) attention.
Compared to dense attention, the benefit of block sparse is to speed up
both training and inference. It could save memory thus support longer
context length.
- [x] Add operator spec and shape inference
- [x] Symbolic shape inference
- [x] Refactor GroupQueryAttention to expose common kernels for kv cache
concatenation, q/k/v transpose etc.
- [x] Add cuda kernel to convert block mask to CSR format
- [x] Add cuda kernel to generate position ids
- [x] Add compile script and template files to convert triton kernel to
cubin and dispatcher.
- [x] Add triton kernel v1 for prompt
- [x] Add triton kernel v2 for token generation and support padding
- [x] Update IO Binding Helper to allow buffer sharing.
- [x] Test relevance
- [x] Test performance
### Performance
Test in A100-SXM4-80GB with `batch_size=4, num_heads=32,
max_seq_len=8192, head_size=128, sparse_block_size=64, local_blocks=16,
vert_stride=8, num_layout=8`
We compare sparse attention to corresponding GQA with local attention
windows size 1024, or GQA with dense causal.
Average latency in milliseconds (for fused attention kernel used in
prompt prefilling):
seq_len | GQA-Dense | GQA-Local | SparseAttention
-- | -- | -- | --
64 | 0.0465 | 0.0722 | 0.0641
128 | 0.0618 | 0.0787 | 0.0672
256 | 0.1086 | 0.1076 | 0.0943
512 | 0.2535 | 0.2487 | 0.1676
1024 | 0.7042 | 0.7050 | 0.3800
2048 | 2.4125 | 1.9316 | 0.8966
4096 | 8.9346 | 4.5699 | 2.1129
8192 | 40.5401 | 10.3508 | 5.1748
Average latency in milliseconds (for fused attention kernel used in
token generation:
past_seq_len | GQA-Dense | GQA-Local | SparseAttention
-- | -- | -- | --
64 | 0.0186 | 0.0186 | 0.0870
128 | 0.0408 | 0.0466 | 0.1165
256 | 0.0530 | 0.0592 | 0.0988
512 | 0.0445| 0.0447 | 0.1150
1024 | 0.0634 | 0.0640 | 0.1454
2048 | 0.1027 | 0.0637 | 0.1589
4096 | 0.1789 | 0.0631 | 0.1806
8192 | 0.3288 | 0.0655 | 0.2146
We can see that the kernel for token generation still have room to
improve.
#### Limitations
Only support right-side padding and unidirectional attention.
The following are not supported in the first version:
(1) Packed mode like PackedMultiHeadAttention where input has been
removed padding.
(2) paged attention.
(3) bidirectional attention.
(4) GPU compute capacity that is not 8.0, 8.6 and 8.9.
(5) Left side padding.
Some of these limitations will be removed in the future (may be in a new
operator).
### Description
Add bf16 support for below ops:
ConstantOfShape
Exp
Erf
convolution
PythonOp
### Motivation and Context
phimm model works on bf16, ORT need support bf16 on previous ops to work
with phimm on bf16
### Description
Support GQA operator on CPU with FP32.
### Motivation and Context
Right now, models generated for CPU and GPU must be different. GQA CPU
allows these models to be the same.
### Description
This PR registers the following opset 20 operators to the DML EP:
-IsNaN-20
-IsInf-20
-ReduceMax-20
### 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. -->
### Description
Add GemmaRotaryEmbedding kernel which includes sin and cos in
GemmaRotaryEmbedding forward and apply_rotary_pos_emb. See
gemma_rotary_emb_impl.cu for subgraph details
### 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. -->
### Description
update with ONNX 1.16.0 branch according to
https://github.com/microsoft/onnxruntime/blob/main/docs/How_To_Update_ONNX_Dev_Notes.md
ONNX 1.16.0 release notes:
https://github.com/onnx/onnx/releases/tag/v1.16.0
#### Updated ops for CPU EP:
- DequantizeLinear(21)
- Added int16 and uint16 support + various optimizer tests
- Missing int4 and uint4 support
- Missing block dequantization support
- QuantizeLinear(21)
- Added int16 and uint16 support + various optimizer tests
- Missing int4 and uint4 support
- Missing block quantization support
- Cast(21)
- Missing int4 and uint4 support
- CastLike(21)
- Missing int4 and uint4 support
- ConstantOfShape(21)
- Missing int4 and uint4 support
- Identity(21)
- Missing int4 and uint4 support
- If(21)
- Missing int4 and uint4 support
- Loop(21)
- Missing int4 and uint4 support
- Reshape(21)
- Missing int4 and uint4 support
- Scan(21)
- Missing int4 and uint4 support
- Shape(21)
- Missing int4 and uint4 support
- Size(21)
- Missing int4 and uint4 support
- Flatten(21)
- Missing float8e4m3fnuz, float8e5m2, float8e5m2fnuz, int4, and uint4
support
- Pad(21)
- Missing float8e4m3fnuz, float8e5m2, float8e5m2fnuz, int4, and uint4
support
- Squeeze(21)
- Missing float8e4m3fnuz, float8e5m2, float8e5m2fnuz, int4, and uint4
support
- Transpose(21)
- Missing float8e4m3fnuz, float8e5m2, float8e5m2fnuz, int4, and uint4
support
- Unsqueeze(21)
- Missing float8e4m3fnuz, float8e5m2, float8e5m2fnuz, int4, and uint4
support
#### Unimplemented opset 21 features/ops
- int4 and uint4 data type
- QLinearMatMul(21)
- GroupNormalization(21)
- ai.onnx.ml.TreeEnsemble(5)
### 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. -->
### Disabled tests
#### ORT Training
orttraining/orttraining/test/python/orttraining_test_ort_apis_py_bindings.py
- test_ort_custom_ops: Potential shape inference bug for custom ops
#### Python quantization unit tests
test/onnx/python/quantization (shape inference bug)
- test_op_conv_transpose.py: test_quantize_conv_transpose_u8u8_fp16
- test_op_conv_transpose.py: test_quantize_conv_transpose_s8s8_fp16
- test_op_gemm.py: test_quantize_qop_gemm_s8s8
- test_op_gemm.py: test_quantize_qop_gemm_e4m3fn_same
- test_op_gemm.py: test_quantize_qop_gemm_e4m3fn_p3
- test_op_matmul.py: test_quantize_matmul_u8u8_f16
- test_op_matmul.py: test_quantize_matmul_s8s8_f16
- test_op_matmul.py: test_quantize_matmul_s8s8_f16_entropy
- test_op_matmul.py: test_quantize_matmul_s8s8_f16_percentile
- test_op_matmul.py: test_quantize_matmul_s8s8_f16_distribution
- test_op_relu.py: test_quantize_qop_relu_s8s8
#### ONNX tests
- test_maxpool_2d_ceil_output_size_reduce_by_one: ONNX 1.16.0 fixed a
maxpool output size bug and added this test. Enable this test when [ORT
PR](https://github.com/microsoft/onnxruntime/pull/18377) is merged.
Refer to original [ONNX PR](https://github.com/onnx/onnx/pull/5741).
- test_ai_onnx_ml_tree_ensemble_set_membership_cpu: new unimplemented op
ai.onnx.ml.TreeEnsemble
- test_ai_onnx_ml_tree_ensemble_single_tree_cpu: same
- test_ai_onnx_ml_tree_ensemble_set_membership_cuda: same
- test_ai_onnx_ml_tree_ensemble_single_tree_cuda: same
- test_cast_INT4_to_FLOAT_cpu: ORT Cast(21) impl doesn't support int4
yet
- test_cast_INT4_to_INT8_cpu: same
- test_cast_UINT4_to_FLOAT_cpu: same
- test_cast_UINT4_to_UINT8_cpu: same
- test_cast_INT4_to_FLOAT_cuda
- test_cast_INT4_to_INT8_cuda
- test_cast_UINT4_to_FLOAT_cuda
- test_cast_UINT4_to_UINT8_cuda
- test_constantofshape_float_ones_cuda: ConstantOfShape(21) not
implemented for cuda
- test_constantofshape_int_shape_zero_cuda: same
- test_constantofshape_int_zeros_cuda: same
- test_flatten_axis0_cuda: Flatten(21) not implemented for cuda
- test_flatten_axis1_cuda: same
- test_flatten_axis2_cuda: same
- test_flatten_axis3_cuda: same
- test_flatten_default_axis_cuda: same
- test_flatten_negative_axis1_cuda: same
- test_flatten_negative_axis2_cuda: same
- test_flatten_negative_axis3_cuda: same
- test_flatten_negative_axis4_cuda: same
- test_qlinearmatmul_2D_int8_float16_cpu: QLinearMatMul(21) for onnx not
implemented in ORT yet
- test_qlinearmatmul_2D_int8_float32_cpu: same
- test_qlinearmatmul_2D_uint8_float16_cpu: same
- test_qlinearmatmul_2D_uint8_float32_cpu: same
- test_qlinearmatmul_3D_int8_float16_cpu: same
- test_qlinearmatmul_3D_int8_float32_cpu: same
- test_qlinearmatmul_3D_uint8_float16_cpu: same
- test_qlinearmatmul_3D_uint8_float32_cpu: same
- test_qlinearmatmul_2D_int8_float16_cuda: same
- test_qlinearmatmul_2D_int8_float32_cuda: same
- test_qlinearmatmul_2D_uint8_float16_cuda: same
- test_qlinearmatmul_2D_uint8_float32_cuda: same
- test_qlinearmatmul_3D_int8_float16_cuda: same
- test_qlinearmatmul_3D_int8_float32_cuda: same
- test_qlinearmatmul_3D_uint8_float16_cuda: same
- test_qlinearmatmul_3D_uint8_float32_cuda: same
- test_size_cuda: Size(21) not implemented for cuda
- test_size_example_cuda: same
- test_dequantizelinear_blocked: Missing implementation for block
dequant for DequantizeLinear(21)
- test_quantizelinear_blocked_asymmetric: Missing implementation for
block quant for QuantizeLinear(21)
- test_quantizelinear_blocked_symmetric: Missing implementation for
block quant for QuantizeLinear(21)
---------
Signed-off-by: liqunfu <liqun.fu@microsoft.com>
Signed-off-by: Ganesan Ramalingam <grama@microsoft.com>
Co-authored-by: Ganesan Ramalingam <grama@microsoft.com>
Co-authored-by: George Wu <jywu@microsoft.com>
Co-authored-by: adrianlizarraga <adlizarraga@microsoft.com>
Although DML doesn't have a "fast" gelu approximation operator, its
standard GELU operator is still faster than having to combine all the
separate elementwise operators from different ops.
### Description
<!-- Describe your changes. -->
1. Introduce latest cutlass extension from TRTLLM that gives us cutlass
upgrade(to 3.4) opportunity from MoE side.
2. Fix Windows build issue
3. Add Int4 MoE op and ut
### 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. -->
### Description
<!-- Describe your changes. -->
1.Support Tensor Parallelism in ShardedMoE.
2.Make necessary code changes to support Mixtral MoE.
3.Fix a bug related to using IOBinding in test script.
4.Fix the input size limitation
### 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. -->
### Description
DML Implementation for [com.microsoft.DynamicQuantizeMatMul
](https://github.com/microsoft/onnxruntime/blob/main/docs/ContribOperators.md#com.microsoft.DynamicQuantizeMatMul)
```
.\onnxruntime_test_all.exe --gtest_filter="*DynamicQuantizeMatMul.*"
Note: Google Test filter = *DynamicQuantizeMatMul.*
[==========] Running 10 tests from 1 test suite.
[----------] Global test environment set-up.
[----------] 10 tests from DynamicQuantizeMatMul
[ RUN ] DynamicQuantizeMatMul.HasZeroPoint_NoBias_test_S8
[ OK ] DynamicQuantizeMatMul.HasZeroPoint_NoBias_test_S8 (635 ms)
[ RUN ] DynamicQuantizeMatMul.HasZeroPoint_NoBias_test_U8
[ OK ] DynamicQuantizeMatMul.HasZeroPoint_NoBias_test_U8 (514 ms)
[ RUN ] DynamicQuantizeMatMul.NoZeroPoint_HasBias_test_S8
[ OK ] DynamicQuantizeMatMul.NoZeroPoint_HasBias_test_S8 (512 ms)
[ RUN ] DynamicQuantizeMatMul.NoZeroPoint_HasBias_test_U8
[ OK ] DynamicQuantizeMatMul.NoZeroPoint_HasBias_test_U8 (505 ms)
[ RUN ] DynamicQuantizeMatMul.NoZeroPoint_NoBias_test_S8
[ OK ] DynamicQuantizeMatMul.NoZeroPoint_NoBias_test_S8 (526 ms)
[ RUN ] DynamicQuantizeMatMul.NoZeroPoint_NoBias_test_U8
[ OK ] DynamicQuantizeMatMul.NoZeroPoint_NoBias_test_U8 (504 ms)
[ RUN ] DynamicQuantizeMatMul.HasZeroPoint_HasBias_test_S8
[ OK ] DynamicQuantizeMatMul.HasZeroPoint_HasBias_test_S8 (512 ms)
[ RUN ] DynamicQuantizeMatMul.HasZeroPoint_HasBias_test_U8
[ OK ] DynamicQuantizeMatMul.HasZeroPoint_HasBias_test_U8 (512 ms)
[ RUN ] DynamicQuantizeMatMul.UInt8_test_with_empty_input
[ OK ] DynamicQuantizeMatMul.UInt8_test_with_empty_input (112 ms)
[ RUN ] DynamicQuantizeMatMul.B_PerColumn_ND
[ OK ] DynamicQuantizeMatMul.B_PerColumn_ND (348 ms)
[----------] 10 tests from DynamicQuantizeMatMul (4685 ms total)
[----------] Global test environment tear-down
[==========] 10 tests from 1 test suite ran. (4686 ms total)
[ PASSED ] 10 tests.
memleakdbg:
----- No memory leaks detected -----
```
### Motivation and Context
- CalculateDynamicQuantizeMatMul to replace CPU EP run reference
- Added more FP32 testcases to isolate all input datatype combinations
---------
Co-authored-by: Xiang Zhang <xianz@microsoft.com>
### Description
Implment IsInf-10,20 for CUDA.
Add FP16 types also on CPU.
### Motivation and Context
Certain models lag in performance due to IsInf not available on CUDA.
### Description
<!-- Describe your changes. -->
1. Support quantized GPTQ weight in huggingface like
[TheBloke/Llama-2-7B-Chat-GPTQ](https://huggingface.co/TheBloke/Llama-2-7B-Chat-GPTQ)
2. Support Act_order for GPTQ
3. Support [HQQ](https://mobiusml.github.io/hqq_blog/) algorithm to
quantize matmul weight and add quant script
### 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. -->
### Description
DML Implementation for
[com.microsoft.MatMulIntegerToFloat](https://github.com/microsoft/onnxruntime/blob/main/docs/ContribOperators.md#com.microsoft.MatMulIntegerToFloat)
```
.\onnxruntime_test_all.exe --gtest_filter="*MatMulIntegerToFloat.*"
Note: Google Test filter = *MatMulIntegerToFloat.*
[==========] Running 22 tests from 1 test suite.
[----------] Global test environment set-up.
[----------] 22 tests from MatMulIntegerToFloat
[ RUN ] MatMulIntegerToFloat.HasZeroPoint_NoBias_test_S8S8
[ OK ] MatMulIntegerToFloat.HasZeroPoint_NoBias_test_S8S8 (620 ms)
[ RUN ] MatMulIntegerToFloat.NoZeroPoint_HasBias_test_S8S8
[ OK ] MatMulIntegerToFloat.NoZeroPoint_HasBias_test_S8S8 (497 ms)
[ RUN ] MatMulIntegerToFloat.NoZeroPoint_NoBias_test_S8S8
[ OK ] MatMulIntegerToFloat.NoZeroPoint_NoBias_test_S8S8 (488 ms)
[ RUN ] MatMulIntegerToFloat.HasZeroPoint_HasBias_test_S8S8
[ OK ] MatMulIntegerToFloat.HasZeroPoint_HasBias_test_S8S8 (503 ms)
[ RUN ] MatMulIntegerToFloat.HasZeroPoint_NoBias_test_U8U8
[ OK ] MatMulIntegerToFloat.HasZeroPoint_NoBias_test_U8U8 (495 ms)
[ RUN ] MatMulIntegerToFloat.NoZeroPoint_HasBias_test_U8U8
[ OK ] MatMulIntegerToFloat.NoZeroPoint_HasBias_test_U8U8 (488 ms)
[ RUN ] MatMulIntegerToFloat.NoZeroPoint_NoBias_test_U8U8
[ OK ] MatMulIntegerToFloat.NoZeroPoint_NoBias_test_U8U8 (492 ms)
[ RUN ] MatMulIntegerToFloat.HasZeroPoint_HasBias_test_U8X8
[ OK ] MatMulIntegerToFloat.HasZeroPoint_HasBias_test_U8X8 (502 ms)
[ RUN ] MatMulIntegerToFloat.HasZeroPoint_NoBias_test_S8U8
[ OK ] MatMulIntegerToFloat.HasZeroPoint_NoBias_test_S8U8 (452 ms)
[ RUN ] MatMulIntegerToFloat.NoZeroPoint_HasBias_test_S8U8
[ OK ] MatMulIntegerToFloat.NoZeroPoint_HasBias_test_S8U8 (454 ms)
[ RUN ] MatMulIntegerToFloat.NoZeroPoint_NoBias_test_S8U8
[ OK ] MatMulIntegerToFloat.NoZeroPoint_NoBias_test_S8U8 (446 ms)
[ RUN ] MatMulIntegerToFloat.HasZeroPoint_HasBias_test_S8U8
[ OK ] MatMulIntegerToFloat.HasZeroPoint_HasBias_test_S8U8 (508 ms)
[ RUN ] MatMulIntegerToFloat.HasZeroPoint_NoBias_test_U8S8
[ OK ] MatMulIntegerToFloat.HasZeroPoint_NoBias_test_U8S8 (456 ms)
[ RUN ] MatMulIntegerToFloat.NoZeroPoint_HasBias_test_U8S8
[ OK ] MatMulIntegerToFloat.NoZeroPoint_HasBias_test_U8S8 (455 ms)
[ RUN ] MatMulIntegerToFloat.NoZeroPoint_NoBias_test_U8S8
[ OK ] MatMulIntegerToFloat.NoZeroPoint_NoBias_test_U8S8 (447 ms)
[ RUN ] MatMulIntegerToFloat.HasZeroPoint_HasBias_test_U8S8
[ OK ] MatMulIntegerToFloat.HasZeroPoint_HasBias_test_U8S8 (465 ms)
[ RUN ] MatMulIntegerToFloat.MatMulIntegerToFloat_FP16_U8U8
[ OK ] MatMulIntegerToFloat.MatMulIntegerToFloat_FP16_U8U8 (111 ms)
[ RUN ] MatMulIntegerToFloat.MatMulIntegerToFloat_FP16_U8S8
[ OK ] MatMulIntegerToFloat.MatMulIntegerToFloat_FP16_U8S8 (115 ms)
[ RUN ] MatMulIntegerToFloat.MatMulIntegerToFloat_FP16_S8S8
[ OK ] MatMulIntegerToFloat.MatMulIntegerToFloat_FP16_S8S8 (114 ms)
[ RUN ] MatMulIntegerToFloat.MatMulIntegerToFloat_FP16_S8U8
[ OK ] MatMulIntegerToFloat.MatMulIntegerToFloat_FP16_S8U8 (110 ms)
[ RUN ] MatMulIntegerToFloat.MatMulIntegerToFloat_FP16
[ OK ] MatMulIntegerToFloat.MatMulIntegerToFloat_FP16 (112 ms)
[ RUN ] MatMulIntegerToFloat.MatMulInteger_With_ZeroPoint
[ OK ] MatMulIntegerToFloat.MatMulInteger_With_ZeroPoint (337 ms)
[----------] 22 tests from MatMulIntegerToFloat (8679 ms total)
[----------] Global test environment tear-down
[==========] 22 tests from 1 test suite ran. (8680 ms total)
[ PASSED ] 22 tests.
memleakdbg:
----- No memory leaks detected -----
```
### 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. -->
* `CalculateMatMulIntegerToFloat` to replace CPU EP run reference
* Added more FP32 testcases to isolate all input datatype combinations
* Added fixed input to `MatMulIntegerToFloat_FP16*` test cases as for
FP16 test cases.
* onnxruntime/test/testdata/matmul_integer_to_float.py` is capable of
generating FP16 models, but we do not produce any for now
I've added NHWC GridSample support to the CUDA EP to reduce the number
of layout transforms. Also I've enabled the full set of GridSampleTests
for all EPs. I've also added the GridSample OpSet 16 to the registered
kernels.
### Motivation and Context
This is the first PR is a series of enhancements of the CUDA EP
improving NHWC support to avoid costly layout transforms between NWHC
and NCHW nodes which are layout sensitive. Also testing was quite
rudimentary for the CUDA EP while it was great for the CPU path. I've
regenerated grid_sample_test.cc enabling tests for other platforms as
well. Those tests resurfaced #10607 again which is fixed as well.
### ONNX Gelu Op in Opset 20
Refactor code to support MSDomain Gelu and ONNX Gelu-opset20 Op
1. Move CPU-GELU implmentation from
`onnxruntime/contrib_ops/cpu/activations.h/cc` to
`onnxruntime/core/providers/cpu/tensor/gelu.h/cc`, as the implementation
for approximate attribute to be 'none'.
2. Dumplicate some logic from
`onnxruntime/contrib_ops/cpu/bert/bias_gelu.cc` to
`onnxruntime/core/providers/cpu/tensor/gelu.h/cc`, as the implementation
for approximate attribute to be 'tanh'.
3. Register ONNX domain Gelu CPU kernel from opset 20 in
`onnxruntime/core/providers/cpu/cpu_execution_provider.cc`.
4. Move `onnxruntime/contrib_ops/cuda/bert/fast_gelu_impl.h/cu` to
`onnxruntime/core/providers/cuda/tensor/gelu_impl.h` and
`onnxruntime/core/providers/cuda/tensor/gelu_approximate_impl.cu`
respectively, as the implementation for approximate attribute to be
'tanh'.
5. Implement the logic for approximate attribute to be 'none' in
`onnxruntime/core/providers/cuda/tensor/gelu_impl.cu`.
6. Register ONNX domain Gelu CUDA kernel from opset 20 in
`onnxruntime/core/providers/cuda/cuda_execution_provider.cc`.
7. ROCM ep related changes.
8. Enrich the tests for ONNX domain Gelu in
`onnxruntime/test/providers/cpu/activation/activation_op_test.cc`.
### Description
Sqrt does not have BF16 support yet. Adding that with this PR
### 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. -->
### Description
<!-- Describe your changes. -->
Adds bfloat16 as a supported dtype for SimplifiedLayerNormFusion which
will provide speedup for Llama-v2 on A100 using bfloat16 numerical
format.
_layernorm_optimized_training.onnx exported in bfloat16 vs. float16:_
### Repro Instructions
```python
from torch import nn
from onnxruntime.training.ortmodule import ORTModule, DebugOptions, LogLevel
import torch
dtype = torch.bfloat16
# dtype = torch.float16
class Net(nn.Module):
def __init__(self):
super().__init__()
self.fc = nn.Linear(784, 10, dtype=dtype)
self.layernorm = nn.LayerNorm([784], dtype=dtype)
def forward(self, x):
x = x.view(x.shape[0], -1)
x = self.layernorm(x)
x = self.fc(x)
return x
model = Net()
model = ORTModule(model, DebugOptions(save_onnx=True, onnx_prefix='layernorm', log_level=LogLevel.INFO))
model.to("cuda")
images = torch.randn((8, 28, 28), dtype=dtype).to("cuda")
output = model(images)
```
### 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. -->
ONNX Runtime integration with Llama-v2 family of LLMs.
---------
Co-authored-by: Prathik Rao <prathikrao@microsoft.com@orttrainingdev8.d32nl1ml4oruzj4qz3bqlggovf.px.internal.cloudapp.net>
`ScatterElements` in opset 18 has been around for a while. However, the
highest opset supporting `ScatterElements` in ORT is 13. This PR
implement this op in CUDA EP by replacing `assignment` in the current
CDUA kernel with `atomic reduction` (e.g., atomic add, atomic max). A
series of fundamental atomic functions (e.g., atomic max for int8_t and
half) are implemented in `common.cuh`; the implementation is general
enough to cover old CUDA and new CUDA versions.
- The core changes are in `cuda/atomic/common.cuh` with very detailed
documentation including `bit-wise operation's visualization`. They are
also copied to `rocm/atomic/common.cuh` to support AMD GPU.
- `/cuda/tensor/gather_elements_impl.cu` contains small changes to call
the new atomic functions to support new `reduction` behavior in new
`ScatterElements`.
- New `ScatterElements` are defined in `rocm_execution_provider.cc` and
`cuda_execution_provider.cc`.
### Description
Implement Pad-18 for Cuda.
### Motivation and Context
Latest models converted by Dynamo fall back on CPU for Pad with
performance degradation.
This contributes to
https://github.com/microsoft/onnx-rewriter/issues/126
### Description
These changes add rotary embedding and packed qkv input to gqa. As of
now, the changes are only supported with Flash-Attention (SM >= 80) but
should soon be supported with Memory Efficient Attention as well.
### Motivation and Context
With the fusion of rotary embedding into this Attention op, we hope to
observe some perf gain. The packed QKV should also provide some perf
gain in the context of certain models, like Llama2, that would benefit
from running ops on the fused QKV matrix, rather than the separate Q, K,
and V.
---------
Co-authored-by: Yufeng Li <liyufeng1987@gmail.com>
### Description
<!-- Describe your changes. -->
Add `temperature` as an input to WhisperBeamSearch op and initialize
correctly in parameter setup.
### Motivation and Context
Currently, temperature is included as an attribute to the BeamSearch op,
which doesn't let the model act dynamically in a single inference
session. By including this variable as an input, the temperature value
can be altered in any inference call (important for 1P teams)
---------
Co-authored-by: Peter McAughan <petermca@microsoft.com>
Co-authored-by: kunal-vaishnavi <115581922+kunal-vaishnavi@users.noreply.github.com>
Co-authored-by: Kunal Vaishnavi <kvaishnavi@microsoft.com>
### Description
<!-- Describe your changes. -->
Register DML operators for opset 19.
- Cast19
- Castlike19
- Constant19
- Equal19
- Identity19
- QuantizeLinear19
- DequantizeLinear19
- Reshape19
- Shape19
- Size
### 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. -->
---------
Co-authored-by: linnealovespie <linneamay@microsoft.com>
### Description
<!-- Describe your changes. -->
1. support causal mask in MHA cpu
2. support custom rotary_dim in rotary_emb
3. add bf16 for rotary_emb
4. fix a bug in attention rotary
### 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. -->
### Description
<!-- Describe your changes. -->
Implements LabelEncoder as per `ai.onnx.ml` opset 4 for the upcoming
ONNX 1.15 release. ~~This currently depends on a new ONNX release
candidate and so is marked as draft in the meantime.~~
### Motivation and Context
Closes https://github.com/microsoft/onnxruntime/issues/17602
### Description
<!-- Describe your changes. -->
### 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. -->
### Description
reducemax/min have been updated in onnx(20). implement it in ort
### Motivation and Context
this is for ort1.17.0 release
---------
Signed-off-by: Liqun Fu <liqfu@microsoft.com>
