reorganize quantization files (#10023)

* reorganize quantization files

onnxruntime/contrib_ops/cpu/quantization/dynamic_quantize_matmul.cc

@@ -5,7 +5,7 @@
 #include "core/mlas/inc/mlas.h"
 #include "core/providers/cpu/math/element_wise_ops.h"
 #include "core/providers/cpu/math/matmul_helper.h"
-#include "core/providers/cpu/math/matmul_integer_base.h"
+#include "core/providers/cpu/quantization/matmul_integer_base.h"
 #include "core/util/math_cpuonly.h"
 #include "core/util/qmath.h"
 

onnxruntime/contrib_ops/cpu/quantization/matmul_integer16.cc

@@ -1,7 +1,7 @@
 // Copyright (c) Microsoft Corporation. All rights reserved.
 // Licensed under the MIT License.
 
-#include "contrib_ops/cpu/matmul_integer16.h"
+#include "matmul_integer16.h"
 #include "core/providers/cpu/math/matmul_helper.h"
 
 namespace onnxruntime {

onnxruntime/contrib_ops/cpu/quantization/qembed_layer_norm.cc

@@ -5,7 +5,7 @@
 
 #include <cmath>
 
-#include "embed_layer_norm_helper.h"
+#include "contrib_ops/cpu/bert/embed_layer_norm_helper.h"
 #include "core/framework/op_kernel.h"
 #include "core/providers/common.h"
 #include "core/quantization/quantization.h"

onnxruntime/contrib_ops/cpu/quantization/qembed_layer_norm.h

@@ -3,7 +3,7 @@
 
 #pragma once
 
-#include "embed_layer_norm.h"
+#include "contrib_ops/cpu/bert/embed_layer_norm.h"
 #include "core/common/common.h"
 #include "core/framework/op_kernel.h"
 

onnxruntime/contrib_ops/cpu/quantization/qlinear_activations.cc

@@ -2,8 +2,8 @@
 // Licensed under the MIT License.
 
 #include "qlinear_activations.h"
-
 #include "qlinear_lookup_table.h"
+
 #include "core/mlas/inc/mlas.h"
 #include "core/platform/threadpool.h"
 

onnxruntime/contrib_ops/cpu/quantization/qlinear_concat.cc

@@ -2,7 +2,8 @@
 // Licensed under the MIT License.
 
 #include "qlinear_concat.h"
-#include "contrib_ops/cpu/qlinear_lookup_table.h"
+#include "qlinear_lookup_table.h"
+
 #include "core/providers/common.h"
 #include "core/mlas/inc/mlas.h"
 #include "core/platform/threadpool.h"

onnxruntime/contrib_ops/cpu/quantization/qlinear_pool.cc

@@ -1,10 +1,10 @@
 // Copyright (c) Microsoft Corporation. All rights reserved.
 // Licensed under the MIT License.
 
+#include "qlinear_global_average_pool.h"
+#include "qlinear_lookup_table.h"
 #include "qlinear_pool.h"
 
-#include "contrib_ops/cpu/qlinear_lookup_table.h"
-#include "contrib_ops/cpu/qlinear_global_average_pool.h"
 #include "core/common/safeint.h"
 #include "core/util/math_cpuonly.h"
 #include "core/providers/common.h"

onnxruntime/contrib_ops/cpu/quantization/quant_gemm.cc

@@ -4,7 +4,7 @@
 #include "core/common/safeint.h"
 #include "core/providers/cpu/math/gemm_base.h"
 #include "core/providers/cpu/math/gemm_helper.h"
-#include "core/providers/cpu/math/matmul_integer_base.h"
+#include "core/providers/cpu/quantization/matmul_integer_base.h"
 #include "core/quantization/quantization.h"
 #include "core/util/math_cpuonly.h"
 

onnxruntime/contrib_ops/cpu/quantization/quantize_ops.cc

@@ -1,7 +1,7 @@
 // Copyright (c) Microsoft Corporation. All rights reserved.
 // Licensed under the MIT License.
 
-#include "core/providers/cpu/tensor/quantize_linear.h"
+#include "core/providers/cpu/quantization/quantize_linear.h"
 #include "core/providers/common.h"
 
 namespace onnxruntime {

onnxruntime/core/providers/cpu/quantization/quantize_linear.cc

@@ -1,7 +1,7 @@
 // Copyright (c) Microsoft Corporation. All rights reserved.
 // Licensed under the MIT License.
 
-#include "core/providers/cpu/tensor/quantize_linear.h"
+#include "core/providers/cpu/quantization/quantize_linear.h"
 #include "core/providers/common.h"
 #include "core/mlas/inc/mlas.h"
 #include "core/util/qmath.h"

onnxruntime/test/providers/cpu/math/quantize_linear_matmul_test.cc

@@ -1,7 +1,7 @@
 // Copyright (c) Microsoft Corporation. All rights reserved.
 // Licensed under the MIT License.
 
-#include "core/providers/cpu/math/quantize_linear_matmul.h"
+#include "core/providers/cpu/quantization/quantize_linear_matmul.h"
 #include "core/mlas/inc/mlas.h"
 
 #include "gtest/gtest.h"