Replies: 1 comment
-
Hi @DingXiangYuanZhiXing, What do you count in the inference time? Only the time to evaluate the user policy? If so, you can do that in plain Python, for example: import time
import ecole
env = ecole.environment.Branching()
instances = ecole.instance.SetCoverGenerator()
for _ in range(10):
obs, action_set, reward_offset, done, info = env.reset(next(instances))
policy_time = 0
while not done:
start = time.perf_counter_ns()
action = policy(obs, action_set)
policy_time += time.perf_counter_ns() - start
obs, action_set, reward, done, info = env.step(action)
# Do something with `policy_time` |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
I know how to record
SolvingTime
in here, but I want to record the inference time instead of total solving time.How to accomplish this?
Beta Was this translation helpful? Give feedback.
All reactions