[tune] Improve PBT example (#4575)

This commit is contained in:
Eric Liang
2019-04-09 20:59:17 -07:00
committed by GitHub
parent dab99d26af
commit f8e8743347
3 changed files with 74 additions and 47 deletions
Binary file not shown.

Before

Width:  |  Height:  |  Size: 32 KiB

After

Width:  |  Height:  |  Size: 50 KiB

+1 -1
View File
@@ -36,7 +36,7 @@ Tune includes a distributed implementation of `Population Based Training (PBT) <
When the PBT scheduler is enabled, each trial variant is treated as a member of the population. Periodically, top-performing trials are checkpointed (this requires your Trainable to support `checkpointing <tune-usage.html#trial-checkpointing>`__). Low-performing trials clone the checkpoints of top performers and perturb the configurations in the hope of discovering an even better variation.
You can run this `toy PBT example <https://github.com/ray-project/ray/blob/master/python/ray/tune/examples/pbt_example.py>`__ to get an idea of how how PBT operates. When training in PBT mode, a single trial may see many different hyperparameters over its lifetime, which is recorded in its ``result.json`` file. The following figure generated by the example shows PBT discovering new hyperparams over the course of a single experiment:
You can run this `toy PBT example <https://github.com/ray-project/ray/blob/master/python/ray/tune/examples/pbt_example.py>`__ to get an idea of how how PBT operates. When training in PBT mode, a single trial may see many different hyperparameters over its lifetime, which is recorded in its ``result.json`` file. The following figure generated by the example shows PBT with optimizing a LR schedule over the course of a single experiment:
.. image:: pbt.png