| .. |
|
compute_optimizer
|
Optimize computation orders (#13672)
|
2022-12-22 15:12:52 +08:00 |
|
qdq_transformer
|
Cjian/where python operator (#12795)
|
2022-12-12 13:27:47 -08:00 |
|
selectors_actions
|
Switch GSL to MS GSL 4.0.0 (#13416)
|
2022-10-29 04:15:20 -07:00 |
|
transpose_optimizer
|
to work with onnx 1.13 rc, implement ver 18 reduce and optioanl ops, … (#13765)
|
2023-01-09 10:26:16 -08:00 |
|
attention_fusion.cc
|
Add ONNX LayerNormalization(17) (#12978)
|
2022-09-23 09:49:27 +10:00 |
|
attention_fusion.h
|
Refactor transformers and other code to reduce memory allocation calls (#10523)
|
2022-02-24 16:17:14 -08:00 |
|
attention_fusion_helper.h
|
Refactor transformers and other code to reduce memory allocation calls (#10523)
|
2022-02-24 16:17:14 -08:00 |
|
bias_dropout_fusion.cc
|
[CUDA] Implement BitmaskDropout, BitmaskBiasDropout and BitmaskDropoutGrad (#11534)
|
2022-05-27 17:24:47 +08:00 |
|
bias_dropout_fusion.h
|
Refactor transformers and other code to reduce memory allocation calls (#10523)
|
2022-02-24 16:17:14 -08:00 |
|
bias_gelu_fusion.cc
|
Refactor transformers and other code to reduce memory allocation calls (#10523)
|
2022-02-24 16:17:14 -08:00 |
|
bias_gelu_fusion.h
|
Refactor transformers and other code to reduce memory allocation calls (#10523)
|
2022-02-24 16:17:14 -08:00 |
|
bias_softmax_fusion.cc
|
Bugfix for BiasSoftmax Fusion (#12517)
|
2022-08-10 07:20:13 +08:00 |
|
bias_softmax_fusion.h
|
Refactor transformers and other code to reduce memory allocation calls (#10523)
|
2022-02-24 16:17:14 -08:00 |
|
cast_elimination.cc
|
|
|
|
cast_elimination.h
|
|
|
|
common_subexpression_elimination.cc
|
Ignore DequantizeLinear nodes in CommonSubexpressionElimination optimizer (#10934)
|
2022-03-23 08:46:01 +10:00 |
|
common_subexpression_elimination.h
|
Refactor transformers and other code to reduce memory allocation calls (#10523)
|
2022-02-24 16:17:14 -08:00 |
|
concat_slice_elimination.cc
|
Refactor transformers and other code to reduce memory allocation calls (#10523)
|
2022-02-24 16:17:14 -08:00 |
|
concat_slice_elimination.h
|
Refactor transformers and other code to reduce memory allocation calls (#10523)
|
2022-02-24 16:17:14 -08:00 |
|
constant_folding.cc
|
Multi-stream execution support (#13495)
|
2022-12-15 07:39:29 -08:00 |
|
constant_folding.h
|
Refactor transformers and other code to reduce memory allocation calls (#10523)
|
2022-02-24 16:17:14 -08:00 |
|
constant_sharing.cc
|
Fix scalar sharing bug (#13299)
|
2022-10-13 11:19:00 +08:00 |
|
constant_sharing.h
|
Share scalar constant initializer (#12878)
|
2022-10-10 13:32:33 +08:00 |
|
conv_activation_fusion.cc
|
Add data type check in ConvAddRelu fusion (#12058)
|
2022-07-01 15:31:15 -07:00 |
|
conv_activation_fusion.h
|
Convert ConvActivationFusion transformer to a selector action transformer. (#10687)
|
2022-03-02 13:47:55 +10:00 |
|
conv_add_act_fusion.cc
|
fuse Conv+Add+activation for CPU from different op-branch (#10987)
|
2022-04-01 09:25:17 +08:00 |
|
conv_add_act_fusion.h
|
fuse Conv+Add+activation for CPU from different op-branch (#10987)
|
2022-04-01 09:25:17 +08:00 |
|
conv_add_fusion.cc
|
|
|
|
conv_add_fusion.h
|
|
|
|
conv_bn_fusion.cc
|
Include opset 15 in Conv+BatchNormalization fusion (#11960)
|
2022-06-27 10:59:14 -07:00 |
|
conv_bn_fusion.h
|
|
|
|
conv_mul_fusion.cc
|
|
|
|
conv_mul_fusion.h
|
|
|
|
div_mul_fusion.cc
|
|
|
|
div_mul_fusion.h
|
|
|
|
dropout_elimination.cc
|
|
|
|
dropout_elimination.h
|
|
|
|
dynamic_quantize_matmul_fusion.cc
|
Refactor transformers and other code to reduce memory allocation calls (#10523)
|
2022-02-24 16:17:14 -08:00 |
|
dynamic_quantize_matmul_fusion.h
|
Refactor transformers and other code to reduce memory allocation calls (#10523)
|
2022-02-24 16:17:14 -08:00 |
|
embed_layer_norm_fusion.cc
|
Switch GSL to MS GSL 4.0.0 (#13416)
|
2022-10-29 04:15:20 -07:00 |
|
embed_layer_norm_fusion.h
|
Refactor transformers and other code to reduce memory allocation calls (#10523)
|
2022-02-24 16:17:14 -08:00 |
|
expand_elimination.cc
|
|
|
|
expand_elimination.h
|
|
|
|
fast_gelu_fusion.cc
|
Add Pow-15 for LayerNorm Fusion and FastGelu Fusion (#12314)
|
2022-07-28 10:23:35 +08:00 |
|
fast_gelu_fusion.h
|
Refactor transformers and other code to reduce memory allocation calls (#10523)
|
2022-02-24 16:17:14 -08:00 |
|
free_dim_override_transformer.cc
|
Ensure graph resolve occurs after free dimension is overridden (#13634)
|
2022-11-14 18:39:29 -08:00 |
|
free_dim_override_transformer.h
|
Switch GSL to MS GSL 4.0.0 (#13416)
|
2022-10-29 04:15:20 -07:00 |
|
gather_fusion.cc
|
Gather to Slice Fusion (#13599)
|
2022-11-10 13:03:30 +08:00 |
|
gather_fusion.h
|
Gather to Slice Fusion (#13599)
|
2022-11-10 13:03:30 +08:00 |
|
gelu_approximation.cc
|
Refactor transformers and other code to reduce memory allocation calls (#10523)
|
2022-02-24 16:17:14 -08:00 |
|
gelu_approximation.h
|
Refactor transformers and other code to reduce memory allocation calls (#10523)
|
2022-02-24 16:17:14 -08:00 |
|
gelu_fusion.cc
|
Refactor transformers and other code to reduce memory allocation calls (#10523)
|
2022-02-24 16:17:14 -08:00 |
|
gelu_fusion.h
|
Refactor transformers and other code to reduce memory allocation calls (#10523)
|
2022-02-24 16:17:14 -08:00 |
|
gemm_activation_fusion.cc
|
Clarify usage of kOnnxDomainAlias. (#10962)
|
2022-03-25 09:52:59 +10:00 |
|
gemm_activation_fusion.h
|
Refactor transformers and other code to reduce memory allocation calls (#10523)
|
2022-02-24 16:17:14 -08:00 |
|
gemm_sum_fusion.cc
|
Fix some static analysis warnings in the core framework (#10033)
|
2021-12-14 14:41:42 -08:00 |
|
gemm_sum_fusion.h
|
Implement a Gemm/Sum fusion pattern (#9699)
|
2021-11-11 18:33:13 -08:00 |
|
gemm_transpose_fusion.cc
|
Register Custom Symbolic of torch.einsum for ORTModule (#9590)
|
2021-11-18 10:13:58 +08:00 |
|
gemm_transpose_fusion.h
|
|
|
|
graph_transformer.cc
|
Enable selector action transformer infrastructure in minimal build. (#8804)
|
2021-08-27 17:16:05 +10:00 |
|
graph_transformer_mgr.cc
|
|
|
|
graph_transformer_mgr.h
|
Refactor transformers and other code to reduce memory allocation calls (#10523)
|
2022-02-24 16:17:14 -08:00 |
|
graph_transformer_utils.cc
|
Consolidate Identical Children Nodes (#14026)
|
2023-01-06 09:03:10 -08:00 |
|
identical_children_consolidation.cc
|
Consolidate Identical Children Nodes (#14026)
|
2023-01-06 09:03:10 -08:00 |
|
identical_children_consolidation.h
|
Consolidate Identical Children Nodes (#14026)
|
2023-01-06 09:03:10 -08:00 |
|
identity_elimination.cc
|
|
|
|
identity_elimination.h
|
|
|
|
initializer.cc
|
Switch GSL to MS GSL 4.0.0 (#13416)
|
2022-10-29 04:15:20 -07:00 |
|
initializer.h
|
[xnnpack] Have Initializer in Mobile related EPs in Minimal_build and creating EP specific dynamic-schema (#12555)
|
2022-09-06 14:32:15 +08:00 |
|
insert_cast_transformer.cc
|
Share scalar constant initializer (#12878)
|
2022-10-10 13:32:33 +08:00 |
|
insert_cast_transformer.h
|
|
|
|
isinf_reducesum_fusion.cc
|
Refactor transformers and other code to reduce memory allocation calls (#10523)
|
2022-02-24 16:17:14 -08:00 |
|
isinf_reducesum_fusion.h
|
Refactor transformers and other code to reduce memory allocation calls (#10523)
|
2022-02-24 16:17:14 -08:00 |
|
layer_norm_fusion.cc
|
New Pattern Support for LayerNormFusion (#14118)
|
2023-01-04 17:51:14 +08:00 |
|
layer_norm_fusion.h
|
[ORTModule] Enable SimplifiedLayerNormalization Fusion (#11580)
|
2022-06-01 15:09:39 +08:00 |
|
matmul_add_fusion.cc
|
|
|
|
matmul_add_fusion.h
|
Refactor transformers and other code to reduce memory allocation calls (#10523)
|
2022-02-24 16:17:14 -08:00 |
|
matmul_integer_to_float.cc
|
Refactor transformers and other code to reduce memory allocation calls (#10523)
|
2022-02-24 16:17:14 -08:00 |
|
matmul_integer_to_float.h
|
Refactor transformers and other code to reduce memory allocation calls (#10523)
|
2022-02-24 16:17:14 -08:00 |
|
matmul_scale_fusion.cc
|
rocblas alt impl during backward pass only (#13352)
|
2022-11-10 00:47:06 +08:00 |
|
matmul_scale_fusion.h
|
Refactor transformers and other code to reduce memory allocation calls (#10523)
|
2022-02-24 16:17:14 -08:00 |
|
matmul_transpose_fusion.cc
|
rocblas alt impl during backward pass only (#13352)
|
2022-11-10 00:47:06 +08:00 |
|
matmul_transpose_fusion.h
|
Refactor transformers and other code to reduce memory allocation calls (#10523)
|
2022-02-24 16:17:14 -08:00 |
|
nchwc_transformer.cc
|
code clean (#12392)
|
2022-08-01 14:12:35 +08:00 |
|
nchwc_transformer.h
|
|
|
|
nhwc_transformer.cc
|
Update kernel matching logic: decouple from op schemas and remove kernel def hashes (#12791)
|
2022-09-20 14:24:59 -07:00 |
|
nhwc_transformer.h
|
make nhwc transformer only apply to CPU ep. (#11882)
|
2022-09-16 18:46:28 +10:00 |
|
noop_elimination.cc
|
Support More Cases in NoOpElimination (#13460)
|
2022-11-01 10:39:52 +08:00 |
|
noop_elimination.h
|
Support More Cases in NoOpElimination (#13460)
|
2022-11-01 10:39:52 +08:00 |
|
not_where_fusion.cc
|
|
|
|
not_where_fusion.h
|
|
|
|
optimizer_execution_frame.cc
|
Update kernel matching logic: decouple from op schemas and remove kernel def hashes (#12791)
|
2022-09-20 14:24:59 -07:00 |
|
optimizer_execution_frame.h
|
Retry Rework execution frame to reduce memory allocations (#11897)
|
2022-06-20 10:29:43 -07:00 |
|
propagate_cast_ops.cc
|
Switch GSL to MS GSL 4.0.0 (#13416)
|
2022-10-29 04:15:20 -07:00 |
|
propagate_cast_ops.h
|
Refactor transformers and other code to reduce memory allocation calls (#10523)
|
2022-02-24 16:17:14 -08:00 |
|
quick_gelu_fusion.cc
|
QuickGelu Fusion (#12417)
|
2022-10-28 18:12:07 +08:00 |
|
quick_gelu_fusion.h
|
QuickGelu Fusion (#12417)
|
2022-10-28 18:12:07 +08:00 |
|
relu_clip_fusion.cc
|
Fix a bug in ReluClip fusion (#9764)
|
2021-11-16 17:59:02 -08:00 |
|
relu_clip_fusion.h
|
|
|
|
reshape_fusion.cc
|
Refactor transformers and other code to reduce memory allocation calls (#10523)
|
2022-02-24 16:17:14 -08:00 |
|
reshape_fusion.h
|
Refactor transformers and other code to reduce memory allocation calls (#10523)
|
2022-02-24 16:17:14 -08:00 |
|
rocm_blas_alt_impl.cc
|
rocblas alt impl during backward pass only (#13352)
|
2022-11-10 00:47:06 +08:00 |
|
rocm_blas_alt_impl.h
|
rocblas alt impl during backward pass only (#13352)
|
2022-11-10 00:47:06 +08:00 |
|
rule_based_graph_transformer.cc
|
Refactor transformers and other code to reduce memory allocation calls (#10523)
|
2022-02-24 16:17:14 -08:00 |
|
skip_layer_norm_fusion.cc
|
Update SkipLayerNorm fusion rules (#13350)
|
2022-10-26 16:15:27 +08:00 |
|
skip_layer_norm_fusion.h
|
Refactor transformers and other code to reduce memory allocation calls (#10523)
|
2022-02-24 16:17:14 -08:00 |
|
slice_elimination.cc
|
Refactor transformers and other code to reduce memory allocation calls (#10523)
|
2022-02-24 16:17:14 -08:00 |
|
slice_elimination.h
|
|
|
|
transformer_memcpy.cc
|
The transformer of memcpy is needed for ROCm EP and MIGraphX EP when fallbacking CPU happens (#10522)
|
2022-02-11 00:53:24 -08:00 |
|
transformer_memcpy.h
|
|
|
|
unsqueeze_elimination.cc
|
Refactor transformers and other code to reduce memory allocation calls (#10523)
|
2022-02-24 16:17:14 -08:00 |
|
unsqueeze_elimination.h
|
|
|
|
utils.cc
|
Enable QDQ cleanup and NHWC optimizers in an extended minimal build. (#10729)
|
2022-03-04 15:45:42 +10:00 |
|
utils.h
|
Share scalar constant initializer (#12878)
|
2022-10-10 13:32:33 +08:00 |
