-
Notifications
You must be signed in to change notification settings - Fork 251
Open
Description
Hi everyone. I'm new to here. I'm developing a framework about computation offloading, which offloads computation-intensive tasks to remote machines.
I want to write a class decorator, which wrap a class to a new class. It acts like the original class but all computation occurs on a remote machine. For example:
class Model:
def inference(self, x):
...
RemoteModel = remote_class_wrapper(Model)
inst = Model()
remote_inst = RemoteModel()
x = inst.inference(123)
y = remote_inst.inference(123)
assert x == y
So my questions are:
- Am I reinventing the wheel?
- Is there room for improvement in my implementation method?
- Do you have any other comments or suggestions?
I'm looking forward to your replies.😉
Metadata
Metadata
Assignees
Labels
No labels