Scott McKay 978c40d853 Make partitioning utils QDQ aware so it does not break up QDQ node units (#19723) ... ### Description <!-- Describe your changes. --> If the EP handles QDQ node units, we need to make sure we do not split those into different partitions. Update the partitioning utils to be QDQ aware. If there are node units we process the logical nodes they represent instead of individual nodes. This ensure we process all nodes in a QDQ node unit at the same time so that they are always in the same partition. ### 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. --> Fix one of the issues in #19590 --------- Co-authored-by: Edward Chen <18449977+edgchen1@users.noreply.github.com>		2024-03-12 10:55:49 +10:00
..
attention	Move attention test data to file (#17158)	2023-08-15 21:31:57 -07:00
cloud_models	CloudEP (#13855)	2023-01-03 10:03:15 -08:00
CNTK	[Linter] Bump ruff and remove pylint (#17797)	2023-10-05 21:07:33 -07:00
custom_execution_provider_library	Prevent GSL_SUPPRESS arguments from being modified by clang-format (#17242)	2023-08-22 18:26:53 -07:00
custom_op_get_const_input_test_library	Allow disable exceptions to work with ort-extensions (#16536)	2023-06-29 18:24:33 -07:00
custom_op_invalid_library	Improve custom op library handle cleanup (#14099)	2023-01-04 17:56:29 -08:00
custom_op_library	Enrich cuda resources with ep options (#19014)	2024-01-11 10:56:07 -08:00
custom_op_local_function	remove use of ai.onnx.ml in test for custom ops and local functions (#19043)	2024-01-10 16:36:50 +01:00
custom_op_openvino_wrapper_library	Run clang-format in CI (#15524)	2023-04-18 09:26:58 -07:00
float8	Fix float 8 rounding on CPU (#16940)	2023-09-07 20:48:25 +02:00
mobilenet_v3_small_excerpt_gen	[CoreML EP] Enable inputs with dynamic shape (#16915)	2023-08-03 18:15:00 -07:00
multi_stream_models	Introduce reused_buffer_index_per_stream in allocation planner which will be reset after computing the reuse buffer for each stream (#19515)	2024-02-22 10:19:08 -08:00
ort_minimal_e2e_test_data	Changes to support standalone custom ops in a minimal build. (#14497)	2023-03-01 11:22:54 +10:00
ort_minimal_test_models
snpe
squeezenet	Update XNNPACK to latest version (#18038)	2023-11-03 09:04:28 -07:00
test_data_generation	Bump ruff linter to 0.2.1 (#19471)	2024-02-08 16:08:27 -08:00
training_api	Introduce a Nominal Checkpoint for On-Device Training (#19232)	2024-01-30 22:11:25 -08:00
transform	[DML] MatrixMultiplyIntegerToFloat (#19608)	2024-03-04 11:55:35 -08:00
transformers	Some changes to Sampling Op (#14218)	2023-01-12 14:15:26 -08:00
TRTEP_test_model
abs_free_dimensions.onnx
add_opset_314159.onnx
alloc_tensor_reuse.onnx
attention_mask1d_fp16.onnx
attention_mask1d_fp32.onnx
attention_mask2d_fp32.onnx
attention_no_mask_fp16.onnx	Flash Attention v2 MHA (#17227)	2023-08-31 13:52:21 -07:00
attention_past_state.onnx
attention_past_state.u8s8.onnx
attention_past_state.u8u8.onnx
bart_tiny.onnx
bert_toy_optimized.onnx
bert_toy_postprocessed.onnx
capi_symbolic_dims.onnx
capi_symbolic_dims.py	Adopt linrtunner as the linting tool - take 2 (#15085)	2023-03-24 15:29:03 -07:00
ckpt_mnist.pt
constant_floats.onnx
conv_autopad.onnx
conv_qdq_s8s8.onnx
conv_qdq_s8s8_perchannel.onnx
conv_qdq_u8u8.onnx
coreml_argmax_cast_test.onnx
coreml_argmax_cast_test.py	Adopt linrtunner as the linting tool - take 2 (#15085)	2023-03-24 15:29:03 -07:00
crop_and_resize.onnx
cuda_graph_with_shape_nodes.onnx	Relax node placement check for CUDA Graph usage (#16358)	2023-06-15 14:03:08 -07:00
custom_op_negpos.onnx
custom_op_single_schema_multi_kernel.onnx	Single-schema-multi-kernel (#15184)	2023-04-27 13:39:59 -07:00
custom_op_string_lower.onnx
custom_op_variadic_io.onnx	Support for custom op variadic inputs/outputs (#13946)	2022-12-23 11:41:15 -08:00
custom_op_variadic_undef_io.onnx	Support for custom op variadic inputs/outputs (#13946)	2022-12-23 11:41:15 -08:00
dynamic_quantize_matmul_int8.onnx
dynamic_quantize_matmul_int8_bias.onnx
dynamic_quantize_matmul_test.py	Adopt linrtunner as the linting tool - take 2 (#15085)	2023-03-24 15:29:03 -07:00
dynamic_quantize_matmul_uint8.onnx
dynamic_quantize_matmul_uint8_bias.onnx
ep_dynamic_graph_input_test.onnx
ep_dynamic_graph_input_test.py	Adopt linrtunner as the linting tool - take 2 (#15085)	2023-03-24 15:29:03 -07:00
ep_partitioning_test_1.onnx
ep_partitioning_test_2.onnx
ep_partitioning_tests.py	Adopt linrtunner as the linting tool - take 2 (#15085)	2023-03-24 15:29:03 -07:00
flatten_broadcast.onnx	Fix CoreML Flatten handling of axis attribute (#16046)	2023-05-24 08:27:32 +10:00
foo_1.onnx	Fix issue with schema lookup where there are custom ops using the ONNX domain (#14492)	2023-02-03 08:05:18 +10:00
foo_1.onnx.ort	Changes to support standalone custom ops in a minimal build. (#14497)	2023-03-01 11:22:54 +10:00
foo_1_clip_11.onnx	Fix issue with schema lookup where there are custom ops using the ONNX domain (#14492)	2023-02-03 08:05:18 +10:00
foo_3.onnx	Fix issue with schema lookup where there are custom ops using the ONNX domain (#14492)	2023-02-03 08:05:18 +10:00
foo_bar_1.onnx	Fix issue with schema lookup where there are custom ops using the ONNX domain (#14492)	2023-02-03 08:05:18 +10:00
foo_bar_2.onnx	Fix issue with schema lookup where there are custom ops using the ONNX domain (#14492)	2023-02-03 08:05:18 +10:00
foo_bar_3.onnx	Fix issue with schema lookup where there are custom ops using the ONNX domain (#14492)	2023-02-03 08:05:18 +10:00
function_opset_test.onnx
function_with_variadics.onnx
fuse_mul_1.onnx
fuse_select_filter.onnx	Versioning for custom op (#18088)	2023-10-31 16:50:27 -07:00
fuse_select_filter_opset_8.onnx	Versioning for custom op (#18088)	2023-10-31 16:50:27 -07:00
gather_with_scalar_indices_then_shape.onnx	[CoreML EP] Add Shape, Gather, and Slice ops (#17153)	2023-08-18 22:34:34 -07:00
gather_with_scalar_indices_then_shape.py	[CoreML EP] Add Shape, Gather, and Slice ops (#17153)	2023-08-18 22:34:34 -07:00
gh_issue_9671.onnx
gh_issue_11717.onnx
identity_9799.onnx
identity_opt.onnx	io_binding to handle optional input of sequence type_proto (#19273)	2024-01-30 10:25:14 -08:00
identity_string.onnx
initializer_as_output.onnx
invalid_dim_param_value_repetition.onnx
issue4829.onnx
LabelEncoder.onnx
layernorm.onnx
layernorm_no_bias.onnx
layout_transform_nonconst_broadcast_input.onnx	Update transpose optimization to be more QDQ aware (#18444)	2023-11-23 08:27:47 +10:00
layout_transform_reshape.onnx	Transpose optimizer enhancements (#15117)	2023-03-28 08:28:17 +10:00
layout_transform_reshape.qdq.onnx	Transpose optimizer enhancements (#15117)	2023-03-28 08:28:17 +10:00
logicaland.onnx
matmul_1.onnx
matmul_2.onnx
matmul_integer_to_float.py	[DML] MatrixMultiplyIntegerToFloat (#19608)	2024-03-04 11:55:35 -08:00
matmul_integer_to_float_int8.onnx	[DML] MatrixMultiplyIntegerToFloat (#19608)	2024-03-04 11:55:35 -08:00
matmul_integer_to_float_int8_bias.onnx	[DML] MatrixMultiplyIntegerToFloat (#19608)	2024-03-04 11:55:35 -08:00
matmul_integer_to_float_int8_int8.onnx	[DML] MatrixMultiplyIntegerToFloat (#19608)	2024-03-04 11:55:35 -08:00
matmul_integer_to_float_int8_int8_bias.onnx	[DML] MatrixMultiplyIntegerToFloat (#19608)	2024-03-04 11:55:35 -08:00
matmul_integer_to_float_uint8.onnx	[DML] MatrixMultiplyIntegerToFloat (#19608)	2024-03-04 11:55:35 -08:00
matmul_integer_to_float_uint8_bias.onnx	[DML] MatrixMultiplyIntegerToFloat (#19608)	2024-03-04 11:55:35 -08:00
matmul_with_dynamic_input_shape.onnx	[CoreML EP] Enable inputs with dynamic shape (#16915)	2023-08-03 18:15:00 -07:00
matmul_with_dynamic_input_shape.py	[CoreML EP] Enable inputs with dynamic shape (#16915)	2023-08-03 18:15:00 -07:00
merge.onnx	Implement lite custom op API (#15778)	2023-05-04 09:49:17 -07:00
mlnet_encoder.onnx
mnist.basic.ort
mnist.basic.v4.ort
mnist.internal_testing_ep.ort
mnist.onnx
mnist.readme.txt
mobilenet_v3_small_excerpt.onnx	[CoreML EP] Enable inputs with dynamic shape (#16915)	2023-08-03 18:15:00 -07:00
mobilenet_v3_small_excerpt_gen.py	[CoreML EP] Enable inputs with dynamic shape (#16915)	2023-08-03 18:15:00 -07:00
model_containing_op_with_function_body.onnx
model_resize_empty_optional_input.onnx
model_with_external_initializer_come_from_user.onnx
model_with_external_initializer_come_from_user.py	Adopt linrtunner as the linting tool - take 2 (#15085)	2023-03-24 15:29:03 -07:00
model_with_external_initializers.onnx
model_with_external_initializers.py	Extend saving models optimized by inference session (#16912)	2023-07-31 16:39:35 -07:00
model_with_invalid_ort_config_json.onnx
model_with_metadata.onnx
model_with_metadata.py	Adopt linrtunner as the linting tool - take 2 (#15085)	2023-03-24 15:29:03 -07:00
model_with_orig_ext_data.bin	Extend saving models optimized by inference session (#16912)	2023-07-31 16:39:35 -07:00
model_with_orig_ext_data.onnx	Extend saving models optimized by inference session (#16912)	2023-07-31 16:39:35 -07:00
model_with_valid_ort_config_json.onnx
mul_1.noopset.onnx
mul_1.onnx
mul_1_dynamic.onnx	cuda graph enhancement (#19636)	2024-03-07 10:15:18 -08:00
mul_16.onnx
nhwc_conv_clip_relu.onnx
nhwc_conv_clip_relu.txt
nhwc_resize_scales_opset11.onnx	[QNN EP] Fix shape inference of NHWC Resize (#15477)	2023-04-20 07:25:25 -07:00
nhwc_resize_scales_opset18.onnx	[QNN EP] Fix shape inference of NHWC Resize (#15477)	2023-04-20 07:25:25 -07:00
nhwc_resize_sizes_opset11.onnx	[QNN EP] Fix shape inference of NHWC Resize (#15477)	2023-04-20 07:25:25 -07:00
nhwc_resize_sizes_opset18.onnx	[QNN EP] Fix shape inference of NHWC Resize (#15477)	2023-04-20 07:25:25 -07:00
nhwc_resize_sizes_opset18.quant.onnx	[QNN EP] Fix shape inference of NHWC Resize (#15477)	2023-04-20 07:25:25 -07:00
nnapi_internal_uint8_support.onnx
nnapi_internal_uint8_support.py	Adopt linrtunner as the linting tool - take 2 (#15085)	2023-03-24 15:29:03 -07:00
nnapi_reshape_flatten_test.onnx
nnapi_reshape_flatten_test.py	Adopt linrtunner as the linting tool - take 2 (#15085)	2023-03-24 15:29:03 -07:00
nnapi_sigmoid_input_rank_test.onnx
nnapi_sigmoid_input_rank_test.py	Adopt linrtunner as the linting tool - take 2 (#15085)	2023-03-24 15:29:03 -07:00
onnx_backend_test_series_filters.jsonc	Label encoder opset4 (#17977)	2024-01-12 12:43:44 -08:00
onnx_backend_test_series_overrides.jsonc	ONNX 1.15 integration (#17125)	2023-09-26 14:44:48 -07:00
optional_1.onnx	Fix issue with schema lookup where there are custom ops using the ONNX domain (#14492)	2023-02-03 08:05:18 +10:00
optional_2.onnx	Implement lite custom op API (#15778)	2023-05-04 09:49:17 -07:00
optional_3.onnx	Implement lite custom op API (#15778)	2023-05-04 09:49:17 -07:00
optional_inputs_ir3.onnx
optional_inputs_ir4.onnx
optional_sequence_tensor.onnx
ort_github_issue_4031.onnx
ort_github_issue_4031.onnx.ort
ort_github_issue_4031.py
ort_github_issue_10305.onnx
ort_github_issue_10305.py
ort_github_issue_11536.onnx
ort_github_issue_12151.onnx
ort_github_issue_15949.onnx	Fix some issues with NNAPI Softmax (#16095)	2023-06-08 13:56:06 +10:00
ort_github_issue_17000.onnx	Two fixes involving minimal builds (#17000)	2023-08-23 16:01:22 +10:00
ort_github_issue_17000.ort	Two fixes involving minimal builds (#17000)	2023-08-23 16:01:22 +10:00
ort_github_issue_17000.py	Two fixes involving minimal builds (#17000)	2023-08-23 16:01:22 +10:00
ort_github_issue_19590.onnx	Make partitioning utils QDQ aware so it does not break up QDQ node units (#19723)	2024-03-12 10:55:49 +10:00
ort_github_issue_19590.py	Make partitioning utils QDQ aware so it does not break up QDQ node units (#19723)	2024-03-12 10:55:49 +10:00
overridable_initializer.onnx
packed_attention_fp16.onnx	Add PackedAttention for packing mode (#14858)	2023-03-21 12:59:29 -07:00
packed_attention_fp16.rbp.onnx	add cutlass fmha support in PackedAttention (#15838)	2023-05-11 13:47:15 -07:00
packed_attention_fp16.rbp.py	add cutlass fmha support in PackedAttention (#15838)	2023-05-11 13:47:15 -07:00
packed_attention_fp32.onnx	Add PackedAttention for packing mode (#14858)	2023-03-21 12:59:29 -07:00
Pads.bin
pipeline_vectorize.onnx
pyop_1.onnx
pyop_2.onnx
pyop_3.onnx
qdq_with_multi_consumer_dq_nodes.onnx
qdq_with_multi_consumer_q_dq_axis.onnx	Fix the duplicated QDQ attributes setup issue (#18039)	2024-01-10 18:36:33 -08:00
qnn_ctx_2_inputs_order_test.onnx	Fix issue that the generated context cache model inputs/outputs order is not guaranteed (#19195)	2024-01-19 15:16:17 -08:00
qnn_ep_partial_support.onnx	[QNN EP] Add session option to disable fallback to default CPU EP (#16016)	2023-05-23 17:56:32 -07:00
reduced_build_test.onnx_model_with_excluded_ops
reduced_build_test.readme.txt
required_ops.config	Two fixes involving minimal builds (#17000)	2023-08-23 16:01:22 +10:00
required_ops_and_types.config	Two fixes involving minimal builds (#17000)	2023-08-23 16:01:22 +10:00
required_ops_config.readme.txt
scan_1.onnx
sequence_construct.onnx
sequence_insert.onnx
sequence_length.onnx
shape_then_slice_and_gather.onnx	[CoreML EP] Add Shape, Gather, and Slice ops (#17153)	2023-08-18 22:34:34 -07:00
shape_then_slice_and_gather.py	[CoreML EP] Add Shape, Gather, and Slice ops (#17153)	2023-08-18 22:34:34 -07:00
sklearn_bin_voting_classifier_soft.onnx
sklearn_bin_voting_classifier_soft.onnx.ort
sklearn_bin_voting_classifier_soft.readme.txt
sparse_initializer_as_output.onnx
sparse_initializer_as_output.py	Bump ruff in CI (#15533)	2023-04-17 10:11:44 -07:00
sparse_to_dense_matmul.onnx
sparse_to_dense_matmul.py	Bump ruff in CI (#15533)	2023-04-17 10:11:44 -07:00
subgraph_implicit_input_from_initializer.onnx
subgraph_input_shadows_outer_scope_value.onnx
test_cast_back_to_back_non_const_mixed_types_origin.onnx
test_conv_follow_convtrans.onnx
test_conv_follow_convtrans_s8.onnx
test_kernel_info_get_const_input.onnx	FasterTransformer model wrapper using custom op (#15013)	2023-03-20 09:05:30 -07:00
test_kernel_info_get_const_input.py	Adopt linrtunner as the linting tool - take 2 (#15085)	2023-03-24 15:29:03 -07:00
test_model_with_fullonnxdomain.onnx
test_resize.onnx
test_training_model.onnx
test_training_model_0.onnx
test_training_model_1.onnx
test_training_model_2.onnx
transpose_optimizer_shared_initializers.onnx	Update transpose optimization to be more QDQ aware (#18444)	2023-11-23 08:27:47 +10:00
transpose_optimizer_shared_initializers.py	Update transpose optimization to be more QDQ aware (#18444)	2023-11-23 08:27:47 +10:00
transpose_optimizer_shared_initializers_broadcast.onnx	Make transpose optimizer able to look past DQ node for const initializer (#17618)	2023-09-27 21:17:16 +10:00
transpose_optimizer_shared_initializers_broadcast2.onnx	Update transpose optimization to be more QDQ aware (#18444)	2023-11-23 08:27:47 +10:00
tree_ensemble_as_tensor.onnx
trt_plugin_custom_op_test.onnx	Add TRT plugins support using custom ops (#13847)	2023-04-18 20:24:32 -07:00
trt_plugin_custom_op_test.py	Add TRT plugins support using custom ops (#13847)	2023-04-18 20:24:32 -07:00
trt_reshape.onnx	[TensorRT EP] Fix bug for shape tensor input (#18253)	2023-11-03 16:07:50 -07:00
trt_reshape_test.py	[TensorRT EP] Fix bug for shape tensor input (#18253)	2023-11-03 16:07:50 -07:00
unused_initializer.onnx
VariedInputCustomOp.onnx
zipmap_int64float.onnx
zipmap_stringfloat.onnx