satyajandhyala 9f69b2f291 Added InsertAndReduce strategy to PropagateCastOps transformation in addition to FloodFill strategy (#7454) ... * Moved GraphTransformerConfiguration to a separate file and added strategy option to PropagateCastOps transformation. * Added testing both FloodFill and InsertAndReduce stratigies for cast propagation. * Added AddConsumer and RemoveConsumer functions to in graph.h for efficient graph editing. * Added PropagateCastOps code documentation * Added GraphTransformationConfiguration class hierarchy information * Added RemoveInputOutputUpDownCasts		2021-05-10 20:46:28 -07:00
..
qdq_transformer	Add QDQ support for MatMulIntegerToFloat, Gather and Transpose (#7500)	2021-05-03 15:51:25 -07:00
attention_fusion.cc	Sync ORTModule branch with master and fix tests (#6526)	2021-02-02 08:59:56 -08:00
attention_fusion.h
attention_fusion_helper.h
bias_dropout_fusion.cc	Fix build, cleanup.	2021-02-17 11:50:41 -08:00
bias_dropout_fusion.h	Fix build, cleanup.	2021-02-17 11:50:41 -08:00
bias_gelu_fusion.cc
bias_gelu_fusion.h
bias_softmax_fusion.cc	Sync ORTModule branch with master and fix tests (#6526)	2021-02-02 08:59:56 -08:00
bias_softmax_fusion.h
cast_elimination.cc
cast_elimination.h
common_subexpression_elimination.cc	Change onnxruntime::make_unique to std::make_unique (#7502)	2021-04-29 17:04:53 -07:00
common_subexpression_elimination.h
computation_reduction.cc	Remove condition from ORT_RETURN_IF[_NOT] macro output. (#6563)	2021-02-05 17:33:29 -08:00
computation_reduction.h
concat_slice_elimination.cc	Sync ORTModule branch with master and fix tests (#6526)	2021-02-02 08:59:56 -08:00
concat_slice_elimination.h	Sync ORTModule branch with master and fix tests (#6526)	2021-02-02 08:59:56 -08:00
constant_folding.cc	Fix regression in constant folding optimizer (#6795)	2021-02-24 19:10:14 -08:00
constant_folding.h	Skip constant folding dequantizelinear for quant qdq format (#6643)	2021-02-11 14:06:13 -08:00
conv_activation_fusion.cc	exclude non support types (#6653)	2021-02-12 13:30:48 -08:00
conv_activation_fusion.h
conv_add_fusion.cc	Sync ORTModule branch with master and fix tests (#6526)	2021-02-02 08:59:56 -08:00
conv_add_fusion.h
conv_bn_fusion.cc	Change onnxruntime::make_unique to std::make_unique (#7502)	2021-04-29 17:04:53 -07:00
conv_bn_fusion.h
conv_mul_fusion.cc	Change onnxruntime::make_unique to std::make_unique (#7502)	2021-04-29 17:04:53 -07:00
conv_mul_fusion.h
div_mul_fusion.cc	Div mul fusion (#7183)	2021-04-05 09:35:30 -07:00
div_mul_fusion.h	Div mul fusion (#7183)	2021-04-05 09:35:30 -07:00
dropout_elimination.cc	Sync ORTModule branch with master and fix tests (#6526)	2021-02-02 08:59:56 -08:00
dropout_elimination.h
dynamic_quantize_matmul_fusion.cc
dynamic_quantize_matmul_fusion.h
embed_layer_norm_fusion.cc	support opset13 in embednorm (#6866)	2021-03-02 12:33:40 -08:00
embed_layer_norm_fusion.h
expand_elimination.cc	Change onnxruntime::make_unique to std::make_unique (#7502)	2021-04-29 17:04:53 -07:00
expand_elimination.h
fast_gelu_fusion.cc	ORTModule - FastGeluFusion/fp16 fix and minor LayerNormFusion cleanup (#6734)	2021-02-18 13:17:16 -08:00
fast_gelu_fusion.h
free_dim_override_transformer.cc
free_dim_override_transformer.h
gelu_approximation.cc	Sync ORTModule branch with master and fix tests (#6526)	2021-02-02 08:59:56 -08:00
gelu_approximation.h
gelu_fusion.cc
gelu_fusion.h
gemm_activation_fusion.cc
gemm_activation_fusion.h
gemm_transpose_fusion.cc	Gemm transpose fusion (#7306)	2021-04-20 09:35:05 -07:00
gemm_transpose_fusion.h	Gemm transpose fusion (#7306)	2021-04-20 09:35:05 -07:00
graph_transformer.cc
graph_transformer_mgr.cc
graph_transformer_mgr.h
graph_transformer_utils.cc	Add ability for memory arenas to "shrink" periodically (#7284)	2021-05-08 07:53:21 -07:00
identity_elimination.cc	Identity elimination with graph output (#7312)	2021-04-19 16:36:35 -07:00
identity_elimination.h
initializer.cc
initializer.h	QDQ: type conversion and more ops support (#7243)	2021-04-06 15:30:31 -07:00
insert_cast_transformer.cc
insert_cast_transformer.h
isinf_reducesum_fusion.cc	IsInf ReduceSum transform (#7188)	2021-04-13 16:05:21 -07:00
isinf_reducesum_fusion.h	IsInf ReduceSum transform (#7188)	2021-04-13 16:05:21 -07:00
layer_norm_fusion.cc	Modify SimplifiedLayerNormFusion to allow fusion in the presence of Casts optionally (#7352)	2021-04-19 19:59:23 -07:00
layer_norm_fusion.h	Modify SimplifiedLayerNormFusion to allow fusion in the presence of Casts optionally (#7352)	2021-04-19 19:59:23 -07:00
matmul_add_fusion.cc
matmul_add_fusion.h
matmul_integer_to_float.cc	Sync ORTModule branch with master and fix tests (#6526)	2021-02-02 08:59:56 -08:00
matmul_integer_to_float.h
matmul_scale_fusion.cc	Add comment (#6860)	2021-03-02 18:54:25 -08:00
matmul_scale_fusion.h
matmul_transpose_fusion.cc	Fix Transpose and MatMul fusion code to check the input datatypes as … (#7147)	2021-03-28 09:24:12 -07:00
matmul_transpose_fusion.h
nchwc_transformer.cc	Implement NCHWc Upsample linear mode (#7623)	2021-05-10 12:16:16 -07:00
nchwc_transformer.h
nhwc_transformer.cc	Change onnxruntime::make_unique to std::make_unique (#7502)	2021-04-29 17:04:53 -07:00
nhwc_transformer.h
not_where_fusion.cc	Not-where fusion (#7182)	2021-04-06 16:12:26 -07:00
not_where_fusion.h	Not-where fusion (#7182)	2021-04-06 16:12:26 -07:00
optimizer_execution_frame.cc	Cleanup a change to ExecutionFrame a little (#7576)	2021-05-06 19:29:34 +10:00
optimizer_execution_frame.h	Cleanup a change to ExecutionFrame a little (#7576)	2021-05-06 19:29:34 +10:00
propagate_cast_ops.cc	Added InsertAndReduce strategy to PropagateCastOps transformation in addition to FloodFill strategy (#7454)	2021-05-10 20:46:28 -07:00
propagate_cast_ops.h	Added InsertAndReduce strategy to PropagateCastOps transformation in addition to FloodFill strategy (#7454)	2021-05-10 20:46:28 -07:00
relu_clip_fusion.cc
relu_clip_fusion.h
reshape_fusion.cc
reshape_fusion.h
rule_based_graph_transformer.cc
shape_to_initializer.cc
shape_to_initializer.h
skip_layer_norm_fusion.cc	Support skiplayernorm fusion without beta in layernorm (#6617)	2021-02-10 17:50:10 -08:00
skip_layer_norm_fusion.h
slice_elimination.cc
slice_elimination.h
transformer_memcpy.cc
transformer_memcpy.h
unsqueeze_elimination.cc	Sync ORTModule branch with master and fix tests (#6526)	2021-02-02 08:59:56 -08:00
unsqueeze_elimination.h
utils.cc	Change onnxruntime::make_unique to std::make_unique (#7502)	2021-04-29 17:04:53 -07:00
utils.h	Propagate QDQ only when scale and zp are scalar (#7492)	2021-04-29 14:40:41 -07:00