def run():
launcher = doodad.AzureMode(azure_subscription_id=AZ_SUB_ID,
azure_storage_connection_str=AZ_CONN_STR,
azure_client_id=AZ_CLIENT_ID,
azure_authentication_key=AZ_SECRET,
azure_tenant_id=AZ_TENANT_ID,
azure_storage_container=AZ_CONTAINER,
log_path='doodad_test_experiment',
region='eastus',
instance_type='Standard_DS1_v2')
az_mount = doodad.MountAzure(
'azure_script_output',
mount_point='/output',
)
local_mount = doodad.MountLocal(local_dir=TESTING_DIR,
mount_point='/data',
output=False)
mounts = [local_mount, az_mount]
doodad.run_command(command='cat /data/secret.txt > /output/secret.txt',
mode=launcher,
mounts=mounts,
verbose=True)
def create_mounts(
code_dirs_to_mount=config.CODE_DIRS_TO_MOUNT,
non_code_dirs_to_mount=config.NON_CODE_DIRS_TO_MOUNT,
remote_mount_configs=config.REMOTE_DIRS_TO_MOUNT,
):
non_code_mounts = [
doodad.MountLocal(**non_code_mapping)
for non_code_mapping in non_code_dirs_to_mount
] + [
doodad.MountRemote(**non_code_mapping)
for non_code_mapping in remote_mount_configs
]
if REPO_DIR not in config.CODE_DIRS_TO_MOUNT:
config.CODE_DIRS_TO_MOUNT.append(REPO_DIR)
code_mounts = [
doodad.MountLocal(local_dir=code_dir, pythonpath=True)
for code_dir in code_dirs_to_mount
]
mounts = code_mounts + non_code_mounts
return mounts
def run():
gcp_mount = doodad.MountGCP(gcp_path='secret_output',
mount_point='/output')
local_mount = doodad.MountLocal(local_dir=TESTING_DIR,
mount_point='/data',
output=False)
mounts = [local_mount, gcp_mount]
launcher = doodad.GCPMode(gcp_bucket=GCP_BUCKET,
gcp_log_path='test_doodad/gcp_test',
gcp_project=GCP_PROJECT,
instance_type='f1-micro',
zone='us-west1-a',
gcp_image=GCP_IMAGE,
gcp_image_project=GCP_PROJECT)
doodad.run_command(command='cat /data/secret.txt > /output/secret.txt',
mode=launcher,
mounts=mounts,
verbose=True)
from doodad.wrappers.sweeper import launcher
import d4rl
import d4rl.flow
import gym
ENVS = [
#'flow-ring-random-v0',
#'flow-ring-controller-v0',
'flow-merge-random-v0',
'flow-merge-controller-v0',
]
mounts = []
mounts.append(
doodad.MountLocal(local_dir='~/code/d4rl',
mount_point='/code/d4rl',
pythonpath=True))
sweeper = launcher.DoodadSweeper(mounts=mounts,
docker_img='justinfu/brac_flow:0.3',
docker_output_dir='/root/tmp/offlinerl',
gcp_bucket_name='justin-doodad',
gcp_image='ubuntu-1804-docker-gpu',
gcp_project='qlearning000')
for env_name in ENVS:
env = gym.make(env_name)
_, dataset = os.path.split(env.dataset_filepath)
dirname, _ = os.path.splitext(dataset)
params = {
import d4rl
import d4rl.flow
import gym
# env_name = 'maze2d-eval-umaze-v1'
env_name = "flow-merge-random-v0"
env = gym.make(env_name)
_, dataset = os.path.split(env.dataset_filepath)
dirname, _ = os.path.splitext(dataset)
mounts = []
# mounts.append(doodad.MountLocal(local_dir='~/code/batch_rl_aviral',
# mount_point='/code/batch_rl_private', pythonpath=True))
mounts.append(
doodad.MountLocal(local_dir="~/code/d4rl",
mount_point="/code/d4rl",
pythonpath=True))
# mounts.append(doodad.MountLocal(local_dir='~/.d4rl/rlkit/%s' % dirname,
# mount_point='/datasets'))
mounts.append(
doodad.MountLocal(local_dir="/data/doodad_results",
mount_point="/root/tmp/offlinerl",
output=True))
gcp_launcher = doodad.GCPMode(
gcp_bucket="justin-doodad",
gcp_log_path="doodad/logs/bear",
gcp_project="qlearning000",
instance_type="n1-standard-1",
zone="us-west1-a",
gcp_image="ubuntu-1804-docker-gpu",
'kitchen-complete-v0',
'kitchen-partial-v0',
'kitchen-mixed-v0',
]
FLOW_ENVS = [
'flow-ring-random-v0',
'flow-ring-controller-v0',
'flow-merge-random-v0',
'flow-merge-controller-v0',
]
ENVS.extend(FLOW_ENVS)
mounts = []
mounts.append(doodad.MountLocal(local_dir='~/code/awr',
mount_point='/code/awr', pythonpath=True))
mounts.append(doodad.MountLocal(local_dir='~/code/d4rl',
mount_point='/code/d4rl', pythonpath=True))
sweeper = launcher.DoodadSweeper(
mounts=mounts,
docker_img='justinfu/awr:0.1',
gcp_bucket_name='justin-doodad',
gcp_image='ubuntu-1804-docker-gpu',
gcp_project='qlearning000'
)
flow_sweeper = launcher.DoodadSweeper(
mounts=mounts,
docker_img='justinfu/awr_flow:0.1',
import doodad
from doodad.wrappers.sweeper import launcher
import d4rl
import d4rl.flow
import gym
#env_name = 'maze2d-eval-umaze-v1'
env_name = 'flow-merge-random-v0'
env = gym.make(env_name)
_, dataset = os.path.split(env.dataset_filepath)
dirname, _ = os.path.splitext(dataset)
mounts = []
#mounts.append(doodad.MountLocal(local_dir='~/code/batch_rl_aviral',
# mount_point='/code/batch_rl_private', pythonpath=True))
mounts.append(doodad.MountLocal(local_dir='~/code/d4rl',
mount_point='/code/d4rl', pythonpath=True))
#mounts.append(doodad.MountLocal(local_dir='~/.d4rl/rlkit/%s' % dirname,
# mount_point='/datasets'))
mounts.append(doodad.MountLocal(local_dir='/data/doodad_results/merge-random',
mount_point='/root/tmp/offlinerl', output=True))
gcp_launcher = doodad.GCPMode(
gcp_bucket='justin-doodad',
gcp_log_path='doodad/logs/bear',
gcp_project='qlearning000',
instance_type='n1-standard-1',
zone='us-west1-a',
gcp_image='ubuntu-1804-docker-gpu',
gcp_image_project='qlearning000'
)
local_launcher = doodad.LocalMode()
from doodad.wrappers.sweeper import launcher
import d4rl
import d4rl.flow
import gym
ENVS = [
'flow-ring-random-v0',
'flow-ring-controller-v0',
'flow-merge-random-v0',
'flow-merge-controller-v0',
]
mounts = []
mounts.append(
doodad.MountLocal(local_dir='~/code/d4rl',
mount_point='/code/d4rl',
pythonpath=True))
sweeper = launcher.DoodadSweeper(mounts=mounts,
docker_img='justinfu/brac_flow:0.3',
docker_output_dir='/root/tmp/offlinerl',
gcp_bucket_name='justin-doodad',
gcp_image='ubuntu-1804-docker-gpu',
gcp_project='qlearning000')
for env_name in ENVS:
env = gym.make(env_name)
_, dataset = os.path.split(env.dataset_filepath)
dirname, _ = os.path.splitext(dataset)
params = {
import d4rl
import d4rl.flow
import gym
#env_name = 'maze2d-eval-umaze-v1'
env_name = 'flow-merge-random-v0'
env = gym.make(env_name)
_, dataset = os.path.split(env.dataset_filepath)
dirname, _ = os.path.splitext(dataset)
mounts = []
#mounts.append(doodad.MountLocal(local_dir='~/code/batch_rl_aviral',
# mount_point='/code/batch_rl_private', pythonpath=True))
mounts.append(
doodad.MountLocal(local_dir='~/code/d4rl',
mount_point='/code/d4rl',
pythonpath=True))
#mounts.append(doodad.MountLocal(local_dir='~/.d4rl/rlkit/%s' % dirname,
# mount_point='/datasets'))
mounts.append(
doodad.MountLocal(local_dir='/data/doodad_results',
mount_point='/root/tmp/offlinerl',
output=True))
gcp_launcher = doodad.GCPMode(gcp_bucket='justin-doodad',
gcp_log_path='doodad/logs/bear',
gcp_project='qlearning000',
instance_type='n1-standard-1',
zone='us-west1-a',
gcp_image='ubuntu-1804-docker-gpu',
gcp_image_project='qlearning000')
from doodad.wrappers.sweeper import launcher
import d4rl
import d4rl.flow
import gym
ENVS = [
"flow-ring-random-v0",
"flow-ring-controller-v0",
"flow-merge-random-v0",
"flow-merge-controller-v0",
]
mounts = []
mounts.append(
doodad.MountLocal(local_dir="~/code/d4rl",
mount_point="/code/d4rl",
pythonpath=True))
sweeper = launcher.DoodadSweeper(
mounts=mounts,
docker_img="justinfu/brac_flow:0.3",
docker_output_dir="/root/tmp/offlinerl",
gcp_bucket_name="justin-doodad",
gcp_image="ubuntu-1804-docker-gpu",
gcp_project="qlearning000",
)
for env_name in ENVS:
env = gym.make(env_name)
_, dataset = os.path.split(env.dataset_filepath)
dirname, _ = os.path.splitext(dataset)
import os
import doodad
from doodad.wrappers.sweeper import launcher
from d4rl import infos
import gym
#env_name = 'maze2d-eval-umaze-v1'
env_name = 'flow-ring-random-v0'
_, dataset = os.path.split(infos.DATASET_URLS[env_name])
dirname, _ = os.path.splitext(dataset)
mounts = []
mounts.append(
doodad.MountLocal(local_dir='~/code/awr',
mount_point='/code/awr',
pythonpath=True,
filter_dir=('data', '.git', 'awr_env')))
mounts.append(
doodad.MountLocal(local_dir='~/code/d4rl',
mount_point='/code/d4rl',
pythonpath=True,
filter_dir=('data', '.git', 'scripts')))
mounts.append(
doodad.MountLocal(local_dir='~/.d4rl/rlkit/%s' % dirname,
mount_point='/datasets'))
mounts.append(
doodad.MountLocal(local_dir='/data/doodad/awr',
mount_point='/data',
output=True))
gcp_launcher = doodad.GCPMode(gcp_bucket='justin-doodad',
import d4rl
import d4rl.flow
import gym
# env_name = 'maze2d-eval-umaze-v1'
env_name = "flow-merge-random-v0"
env = gym.make(env_name)
_, dataset = os.path.split(env.dataset_filepath)
dirname, _ = os.path.splitext(dataset)
mounts = []
# mounts.append(doodad.MountLocal(local_dir='~/code/batch_rl_aviral',
# mount_point='/code/batch_rl_private', pythonpath=True))
mounts.append(
doodad.MountLocal(local_dir="~/code/d4rl",
mount_point="/code/d4rl",
pythonpath=True))
# mounts.append(doodad.MountLocal(local_dir='~/.d4rl/rlkit/%s' % dirname,
# mount_point='/datasets'))
mounts.append(
doodad.MountLocal(
local_dir="/data/doodad_results/merge-random",
mount_point="/root/tmp/offlinerl",
output=True,
))
gcp_launcher = doodad.GCPMode(
gcp_bucket="justin-doodad",
gcp_log_path="doodad/logs/bear",
gcp_project="qlearning000",
instance_type="n1-standard-1",