gistfile1.txt

# An unique identifier for the head node and workers of this cluster.
cluster_name: basic

# The maximum number of workers nodes to launch in addition to the head
# node. This takes precedence over min_workers. min_workers default to 0.
min_workers: 1
initial_workers: 1
max_workers: 1
autoscaling_mode: aggressive

target_utilization_fraction: 0.9

# If a node is idle for this many minutes, it will be removed.
idle_timeout_minutes: 20

docker:
    image: "rayproject/ray:nightly-cpu" # You can change this to latest-cpu if you don't need GPU support and want a faster startup
    # image: "rayproject/ray:1.5.0-cpu"
    container_name: "ray_container"
    # If true, pulls latest version of image. Otherwise, `docker run` will only pull the image
    # if no cached version is present.
    pull_before_run: True
#    run_options: []  # Extra options to pass into "docker run"

# Cloud-provider specific configuration.
provider:
    type: aws
    region: us-west-2
    availability_zone: us-west-2a
    # cache_stopped_nodes: False

# How Ray will authenticate with newly launched nodes.
auth:
    ssh_user: ubuntu
    ssh_private_key: ~/Documents/aws_secrets/xwjiang-test.pem
head_node:
    InstanceType: m5.xlarge
    ImageId: latest_dlami
    KeyName: xwjiang-test
    # InstanceMarketOptions:
    #     MarketType: spot
           # SpotOptions:
           #     MaxPrice: "9.0"


worker_nodes:
    InstanceType: m5.xlarge
    ImageId: latest_dlami
    KeyName: xwjiang-test
    # # Run workers on spot by default. Comment this out to use on-demand.
    # InstanceMarketOptions:
    #     MarketType: spot
        # SpotOptions:
        #     MaxPrice: "9.0"

file_mounts: {
    /home/ray/anaconda3/lib/python3.7/site-packages/ray/tune: /Users/xwjiang/ray/python/ray/tune,
}

initialization_commands: []
setup_commands:
  - pip install -U boto3
  - pip install ipdb # ray[rllib] #torch torchvision
# Custom commands that will be run on the head node after common setup.
head_setup_commands: []

# Custom commands that will be run on worker nodes after common setup.
worker_setup_commands: []

# # Command to start ray on the head node. You don't need to change this.
head_start_ray_commands:
    - ray stop
    - unset RAY_OVERRIDE_RESOURCES && ray start --head --num-cpus=1 --port=6379 --object-manager-port=8076 --autoscaling-config=~/ray_bootstrap_config.yaml --object-store-memory=1000000000

# Command to start ray on worker nodes. You don't need to change this.
worker_start_ray_commands:
    - ray stop
    - unset RAY_OVERRIDE_RESOURCES && ray start --address=$RAY_HEAD_IP:6379 --object-manager-port=8076 --object-store-memory=1000000000