How to deploy a model after the auto-tuning

after doing the auto-tuning for the network model, how to use the best result to generate the .so for deploy.

with autotvm.apply_history_best(log_file):
with relay.build_config(opt_level=3):
graph, lib, params = relay.build_module.build( net, target=target, params=params)

how to use the lib, params to generate the .so like libtvm_runtime.so, that can be called in C++ program?
Thanks a lot!

2 Likes