giuseros commented on a change in pull request #7785:
URL: https://github.com/apache/tvm/pull/7785#discussion_r615040089
##########
File path: python/tvm/micro/model_library_format.py
##########
@@ -126,20 +125,25 @@ def export_model_library_format(mod:
graph_executor_factory.GraphExecutorFactory
Parameters
----------
- mod : tvm.relay.backend.graph_executor_factory.GraphExecutorFactoryModule
+ mod : tvm.relay.backend.executor_factory.ExecutorFactoryModule
The return value of tvm.relay.build, which will be exported into Model
Library Format.
file_name : str
Path to the .tar archive to generate.
"""
tempdir = utils.tempdir()
+ is_aot = isinstance(mod, executor_factory.AOTExecutorFactoryModule)
Review comment:
Can you help me understand here? I need the `is_aot` logic, to avoid
generating the memory path and to select the correct runtime. Either I add a
`get_factory_type` from the factory or I inspect its type here
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
