No logs using custom model

RLlib
1

I use this the torch vision example custom_fast_model, but there are no logs about training information. How to show the training infomations?

== Status ==
Memory usage on this node: 1.2/9.3 GiB
Using FIFO scheduling algorithm.
Resources requested: 0/1 CPUs, 0/0 GPUs, 0.0/5.22 GiB heap, 0.0/2.61 GiB objects
Result logdir: /home/yan/ray_results/IMPALA
Number of trials: 1/1 (1 PENDING)


== Status ==
Memory usage on this node: 1.2/9.3 GiB
Using FIFO scheduling algorithm.
Resources requested: 0/1 CPUs, 0/0 GPUs, 0.0/5.22 GiB heap, 0.0/2.61 GiB objects
Result logdir: /home/yan/ray_results/IMPALA
Number of trials: 1/1 (1 PENDING)
2

Hey @bug404 , you can set verbosity=2 (or 3) in the call to tune.run() to see more information per iteration.

There is also the possibility to define your own custom loggers in Tune:
https://docs.ray.io/en/master/tune/api_docs/logging.html

1 Like
3

Yeah, it works, thank you very much.