MLproject

name: deep-q

# Can have a docker_env instead of a conda_env, e.g.
docker_env:
  image: betheredge/gym_dev
  environment: ["AWS_ACCESS_KEY_ID", "AWS_SECRET_ACCESS_KEY", "MLFLOW_S3_ENDPOINT_URL", "MLFLOW_TRACKING_URI"]

entry_points:
  main:
    parameters:
      environment: {type: string, default: "Breakout-v4"}
      random_choice_min_rate: {type: float, default: 0.01}
      sample_size: {type: int, default: 32}
      verbose: {type: int, default: 1}
      max_episodes: {type: int, default: 999999999}
      max_steps: {type: int, default: 40000000}
      name_prefix: {type: string, default: ""}
      window: {type: int, default: 4}
      target_network_interval: {type: int, default: 32000}
      start_length: {type: int, default:  200000}
      end_length: {type: int, default: 1000000}
      random_decay_end: {type: int, default: 4000000}

      # Agent Args
      frame_skip: {type: int, default: 4}

    command: "python src/main.py 
              --environment {environment} 
              --random_choice_min_rate {random_choice_min_rate}
              --sample_size {sample_size}
              --verbose {verbose}
              --max_episodes {max_episodes}
              --max_steps {max_steps}
              --name_prefix {name_prefix}
              --window {window}
              --target_network_interval {target_network_interval}
              --start_length {start_length}
              --end_length {end_length}
              --random_decay_end {random_decay_end}
              --frame_skip {frame_skip}
              --duel
              --double
              --clip_reward
              --prio"
    all:
      worker: {type: int, default: 2}
      commmand: