Skip to content

[rllib] PPO doesn't work with fractional num gpus #3396

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
ericl merged 2 commits into from
Nov 27, 2018
Merged

[rllib] PPO doesn't work with fractional num gpus #3396

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
96599c7
frac ppo
ericl Nov 25, 2018
254b8c4
gpu test
ericl Nov 25, 2018
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
5 changes: 4 additions & 1 deletion python/ray/rllib/optimizers/multi_gpu_optimizer.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,6 +3,7 @@
from __future__ import print_function

import logging
import math
import numpy as np
from collections import defaultdict
import tensorflow as tf
Expand Down Expand Up @@ -44,7 +45,9 @@ def _init(self,
if not num_gpus:
self.devices = ["/cpu:0"]
else:
self.devices = ["/gpu:{}".format(i) for i in range(num_gpus)]
self.devices = [
"/gpu:{}".format(i) for i in range(int(math.ceil(num_gpus)))
Copy link
Contributor

@richardliaw richardliaw Nov 25, 2018

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

isn't numpy already imported?

Copy link
Contributor Author

@ericl ericl Nov 25, 2018

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Better to avoid numpy if you can use standard libraries instead.

]
self.batch_size = int(sgd_batch_size / len(self.devices)) * len(
self.devices)
assert self.batch_size % len(self.devices) == 0
Expand Down
8 changes: 8 additions & 0 deletions test/jenkins_tests/run_multi_node_tests.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -53,6 +53,14 @@ docker run --rm --shm-size=10G --memory=10G $DOCKER_SHA \
--stop '{"training_iteration": 2}' \
--config '{"simple_optimizer": true, "num_sgd_iter": 2, "model": {"use_lstm": true}}'

docker run --rm --shm-size=10G --memory=10G $DOCKER_SHA \
python /ray/python/ray/rllib/train.py \
--env CartPole-v1 \
--run PPO \
--stop '{"training_iteration": 2}' \
--config '{"num_gpus": 0.1}' \
--ray-num-gpus 1

docker run --rm --shm-size=10G --memory=10G $DOCKER_SHA \
python /ray/python/ray/rllib/train.py \
--env CartPole-v1 \
Expand Down