From 5a7178d3b16c20a3fd9aa11f91f5c9fe7559a493 Mon Sep 17 00:00:00 2001 From: Vihang Mehta Date: Mon, 18 Nov 2024 16:45:40 -0800 Subject: [PATCH] Fix transformations Signed-off-by: Vihang Mehta --- .../src/transformations/transformation_pipeline.cpp | 2 +- .../intel_gpu/src/plugin/transformations_pipeline.cpp | 6 +++--- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/src/plugins/intel_cpu/src/transformations/transformation_pipeline.cpp b/src/plugins/intel_cpu/src/transformations/transformation_pipeline.cpp index 9bac8179105f7a..313b7141dd4fce 100644 --- a/src/plugins/intel_cpu/src/transformations/transformation_pipeline.cpp +++ b/src/plugins/intel_cpu/src/transformations/transformation_pipeline.cpp @@ -665,7 +665,7 @@ void Transformations::PreLpt(const std::vector& defaultPrecis CPU_SET_CALLBACK_COMMON(manager, [&defaultPrecisions](const_node_ptr &node) -> bool { - return ov::pass::low_precision::NetworkHelper::areQuantizeAndDequantizeSupportedForMultiply(node, defaultPrecisions); + return !ov::pass::low_precision::NetworkHelper::areQuantizeAndDequantizeSupportedForMultiply(node, defaultPrecisions); }, ov::pass::ConvertQuantizeDequantize); } diff --git a/src/plugins/intel_gpu/src/plugin/transformations_pipeline.cpp b/src/plugins/intel_gpu/src/plugin/transformations_pipeline.cpp index 706689966cd058..8dbf3c8776bf7d 100644 --- a/src/plugins/intel_gpu/src/plugin/transformations_pipeline.cpp +++ b/src/plugins/intel_gpu/src/plugin/transformations_pipeline.cpp @@ -310,6 +310,8 @@ void TransformationsPipeline::apply(std::shared_ptr func) { if (!is_model_quantized) pass_config->set_callback(is_non_supported_decompression_op); + manager.register_pass(); + const bool keep_precision_sensitive_in_fp32_1 = true; const bool convert_input_output_precision = false; const bool store_original_precision_as_rt_attribute = true; @@ -319,8 +321,6 @@ void TransformationsPipeline::apply(std::shared_ptr func) { convert_input_output_precision, store_original_precision_as_rt_attribute); - manager.register_pass(); - pass_config->set_callback([&](const std::shared_ptr node){ GPU_DEBUG_IF(cldnn::debug_configuration::get_instance()->enable_sdpa != -1) { GPU_DEBUG_CODE(return cldnn::debug_configuration::get_instance()->enable_sdpa == 1); @@ -638,7 +638,7 @@ void TransformationsPipeline::apply(std::shared_ptr func) { if (enableInt8) { pass_config->set_callback([&](const_node_ptr &node) -> bool { - return ov::pass::low_precision::NetworkHelper::areQuantizeAndDequantizeSupportedForMultiply(node, defaultPrecisions); + return !ov::pass::low_precision::NetworkHelper::areQuantizeAndDequantizeSupportedForMultiply(node, defaultPrecisions); }); }