Why when I run finished my ray tune, and i want to load it back using the following:
tune.Tuner.restore("~/ray_result/some_path", trainable_with_cpu_gpu)
But the results keep asking me to do a tuner.fit().
So when i put tuner.fit()
It keep stuck in a loop of Paused, in the following:
Using HyperBand: num_stopped=0 total_brackets=6
Round #0:
Bracket(Max Size (n)=48, Milestone (r)=2, completed=0.0%): {TERMINATED: 48}
Bracket(Max Size (n)=12, Milestone (r)=16, completed=0.0%): {TERMINATED: 12}
Bracket(Max Size (n)=4, Milestone (r)=100, completed=0.0%): {PAUSED: 4}
Round #1:
Bracket(Max Size (n)=48, Milestone (r)=2, completed=0.0%): {PAUSED: 36}
Why is this the case?