rename to dynamo.convert_module_to_trt_engine

Author: zewenli98

py/torch_tensorrt/_compile.py

@@ -319,7 +319,7 @@ def convert_method_to_trt_engine(
             "convert_method_to_trt_engine call is not supported for ir=fx"
         )
     elif target_ir == _IRType.dynamo:
-        return torch_tensorrt.dynamo.convert_method_to_trt_engine(  # type: ignore[no-any-return]
+        return torch_tensorrt.dynamo.convert_module_to_trt_engine(  # type: ignore[no-any-return]
             module,
             inputs=inputs,
             method_name=method_name,

py/torch_tensorrt/dynamo/__init__.py

@@ -7,7 +7,7 @@
 logger = logging.getLogger(__name__)
 
 if version.parse(sanitized_torch_version()) >= version.parse("2.1.dev"):
-    from ._compiler import compile, convert_method_to_trt_engine
+    from ._compiler import compile, convert_module_to_trt_engine
     from ._exporter import export
     from ._settings import CompilationSettings
     from ._SourceIR import SourceIR

py/torch_tensorrt/dynamo/_compiler.py

@@ -448,7 +448,7 @@ def compile_module(
     return partitioned_module
 
 
-def convert_method_to_trt_engine(
+def convert_module_to_trt_engine(
     module: torch.fx.GraphModule,
     method_name: str = "forward",
     inputs: Optional[Sequence[Input | torch.Tensor]] = None,

tests/py/dynamo/runtime/test_convert_method_to_trt_engine.py

@@ -21,7 +21,7 @@ def forward(self, a, b):
         symbolic_traced_gm = torch.fx.symbolic_trace(model)
 
         # Convert to TensorRT engine
-        trt_engine_str = torch_tensorrt.dynamo.convert_method_to_trt_engine(
+        trt_engine_str = torch_tensorrt.dynamo.convert_module_to_trt_engine(
             symbolic_traced_gm, "forward", inputs=[input_data_0, input_data_1]
         )