feat: Add converter for reflection pad 1d and 2d operation

Signed-off-by: Dheeraj Peri <[email protected]>

Author: peri044

README.md

@@ -48,7 +48,7 @@ If you would like to build outside a docker container, please follow the section
 // Set input datatypes. Allowerd options torch::{kFloat, kHalf, kChar, kInt32, kBool}
 // Size of input_dtypes should match number of inputs to the network.
 // If input_dtypes is not set, default precision follows traditional PyT / TRT rules
-auto input = torch_tensorrt::Input(dims, torch::kHalf)
+auto input = torch_tensorrt::Input(dims, torch::kHalf);
 auto compile_settings = torch_tensorrt::ts::CompileSpec({input});
 // FP16 execution
 compile_settings.enabled_precisions = {torch::kHalf};

core/conversion/converters/BUILD

@@ -71,6 +71,7 @@ cc_library(
         "impl/pooling.cpp",
         "impl/quantization.cpp",
         "impl/reduce.cpp",
+        "impl/reflection_pad.cpp",
         "impl/replication_pad.cpp",
         "impl/select.cpp",
         "impl/shuffle.cpp",

tests/core/conversion/converters/BUILD

@@ -83,6 +83,10 @@ converter_test(
     name = "test_reduce",
 )
 
+converter_test(
+    name = "test_reflection_pad",
+)
+
 converter_test(
     name = "test_replication_pad",
 )