defaults:
  - env: brax
  - experiment_overrides: default
  - trial_spec: default
  - _self_

hyperparameters:
    lr: 3e-4
    gamma: 0.99
    lmbda: 0.95
    clip_ratio: 0.2
    value_coef: 0.5
    entropy_coef: 0.0
    total_time_steps: 50_000_000
    num_steps: 64
    num_mini_batches: 32
    num_envs: 2048
    num_epochs: 16
    max_grad_norm: 0.5
    normalize_advantages: True
    normalize_env: True
    anneal_lr: False
    num_eval: 20
    max_episode_steps: 1000
name: "ppo"
tags: ["ppo_baseline_retuned"]
seed: 0
num_seeds: 1
tune: false
checkpoint_dir: null
trials: 8 
wandb:
  mode: "online" # set to online to activate wandb
  entity: "viper_svg"
  project: "online_sac"