mirror of
https://github.com/wassname/ray.git
synced 2026-07-20 12:40:20 +08:00
Philipp MoritzandRobert Nishihara
791bee343f
* make information available for GAE * buggy version of GAE estimator * fix * add more logging and reweight losses * fix logging * fix loss * adapt advantage calculation * update gae * standardize returns * don't normalize td lambda ret * fix * don't standardize advantages * do standardization earlier * different standardization * initializer * drop into the debugger * fix tensorflow broadcasting bug * vf clipping * don't standardize tdlambdaret * different standardization * use huber loss for value function * refactor -- first half * it runs * fix * update * documentation * linting and tests * fix linting * naming * fix * linting * fix * remove prefix madness * fixes * fix * add value function example * fix linting * remove newline
100 lines
3.6 KiB
Bash
Executable File
100 lines
3.6 KiB
Bash
Executable File
#!/usr/bin/env bash
|
|
|
|
# Cause the script to exit if a single command fails.
|
|
set -e
|
|
|
|
# Show explicitly which commands are currently running.
|
|
set -x
|
|
|
|
ROOT_DIR=$(cd "$(dirname "${BASH_SOURCE:-$0}")"; pwd)
|
|
|
|
DOCKER_SHA=$($ROOT_DIR/../../build-docker.sh --output-sha --no-cache)
|
|
echo "Using Docker image" $DOCKER_SHA
|
|
|
|
python $ROOT_DIR/multi_node_docker_test.py \
|
|
--docker-image=$DOCKER_SHA \
|
|
--num-nodes=5 \
|
|
--num-redis-shards=10 \
|
|
--test-script=/ray/test/jenkins_tests/multi_node_tests/test_0.py
|
|
|
|
python $ROOT_DIR/multi_node_docker_test.py \
|
|
--docker-image=$DOCKER_SHA \
|
|
--num-nodes=5 \
|
|
--num-redis-shards=5 \
|
|
--num-gpus=0,1,2,3,4 \
|
|
--num-drivers=7 \
|
|
--driver-locations=0,1,0,1,2,3,4 \
|
|
--test-script=/ray/test/jenkins_tests/multi_node_tests/remove_driver_test.py
|
|
|
|
python $ROOT_DIR/multi_node_docker_test.py \
|
|
--docker-image=$DOCKER_SHA \
|
|
--num-nodes=5 \
|
|
--num-redis-shards=2 \
|
|
--num-gpus=0,0,5,6,50 \
|
|
--num-drivers=100 \
|
|
--test-script=/ray/test/jenkins_tests/multi_node_tests/many_drivers_test.py
|
|
|
|
python $ROOT_DIR/multi_node_docker_test.py \
|
|
--docker-image=$DOCKER_SHA \
|
|
--num-nodes=1 \
|
|
--mem-size=60G \
|
|
--shm-size=60G \
|
|
--test-script=/ray/test/jenkins_tests/multi_node_tests/large_memory_test.py
|
|
|
|
# Test that the example applications run.
|
|
|
|
# docker run --shm-size=10G --memory=10G $DOCKER_SHA \
|
|
# python /ray/examples/lbfgs/driver.py
|
|
|
|
# docker run --shm-size=10G --memory=10G $DOCKER_SHA \
|
|
# python /ray/examples/rl_pong/driver.py \
|
|
# --iterations=3
|
|
|
|
# docker run --shm-size=10G --memory=10G $DOCKER_SHA \
|
|
# python /ray/examples/hyperopt/hyperopt_simple.py
|
|
|
|
# docker run --shm-size=10G --memory=10G $DOCKER_SHA \
|
|
# python /ray/examples/hyperopt/hyperopt_adaptive.py
|
|
|
|
docker run --shm-size=10G --memory=10G $DOCKER_SHA \
|
|
python /ray/python/ray/rllib/train.py \
|
|
--env PongDeterministic-v0 \
|
|
--alg A3C \
|
|
--num-iterations 2 \
|
|
--config '{"num_workers": 16}'
|
|
|
|
docker run --shm-size=10G --memory=10G $DOCKER_SHA \
|
|
python /ray/python/ray/rllib/train.py \
|
|
--env CartPole-v1 \
|
|
--alg PolicyGradient \
|
|
--num-iterations 2 \
|
|
--config '{"kl_coeff": 1.0, "num_sgd_iter": 10, "sgd_stepsize": 1e-4, "sgd_batchsize": 64, "timesteps_per_batch": 2000, "num_agents": 1}'
|
|
|
|
docker run --shm-size=10G --memory=10G $DOCKER_SHA \
|
|
python /ray/python/ray/rllib/train.py \
|
|
--env CartPole-v1 \
|
|
--alg PolicyGradient \
|
|
--num-iterations 2 \
|
|
--config '{"kl_coeff": 1.0, "num_sgd_iter": 10, "sgd_stepsize": 1e-4, "sgd_batchsize": 64, "timesteps_per_batch": 2000, "num_agents": 1, "use_gae": false}'
|
|
|
|
docker run --shm-size=10G --memory=10G $DOCKER_SHA \
|
|
python /ray/python/ray/rllib/train.py \
|
|
--env Pendulum-v0 \
|
|
--alg EvolutionStrategies \
|
|
--num-iterations 2 \
|
|
--config '{"stepsize": 0.01}'
|
|
|
|
docker run --shm-size=10G --memory=10G $DOCKER_SHA \
|
|
python /ray/python/ray/rllib/train.py \
|
|
--env CartPole-v0 \
|
|
--alg DQN \
|
|
--num-iterations 2 \
|
|
--config '{"lr": 1e-3, "schedule_max_timesteps": 100000, "exploration_fraction": 0.1, "exploration_final_eps": 0.02, "dueling": false, "hiddens": [], "model_config": {"fcnet_hiddens": [64], "fcnet_activation": "relu"}}'
|
|
|
|
docker run --shm-size=10G --memory=10G $DOCKER_SHA \
|
|
python /ray/python/ray/rllib/train.py \
|
|
--env PongNoFrameskip-v4 \
|
|
--alg DQN \
|
|
--num-iterations 2 \
|
|
--config '{"lr": 1e-4, "schedule_max_timesteps": 2000000, "buffer_size": 10000, "exploration_fraction": 0.1, "exploration_final_eps": 0.01, "train_freq": 4, "learning_starts": 10000, "target_network_update_freq": 1000, "gamma": 0.99, "prioritized_replay": true}'
|