146 lines
		
	
	
		
			2.0 KiB
		
	
	
	
		
			YAML
		
	
	
	
	
	
			
		
		
	
	
			146 lines
		
	
	
		
			2.0 KiB
		
	
	
	
		
			YAML
		
	
	
	
	
	
name: DEFAULT
 | 
						|
 | 
						|
vars:
 | 
						|
  foo: bar
 | 
						|
 | 
						|
slurm:
 | 
						|
  name: PerlinExp_Slate_DR# '{config[name]}'
 | 
						|
  partitions:
 | 
						|
    - SINGLE
 | 
						|
  num_parallel_jobs: 50
 | 
						|
  cpus_per_task: 1
 | 
						|
  mem-per-cpu: 3000
 | 
						|
  time: 1440 # in minutes
 | 
						|
  ntasks: 1
 | 
						|
  sh_lines: ['']
 | 
						|
 | 
						|
runner: sb3
 | 
						|
 | 
						|
scheduler:
 | 
						|
  total_timesteps: 10000
 | 
						|
  repetitions: 3
 | 
						|
  agents_per_job: 3
 | 
						|
  reps_per_agent: 1
 | 
						|
  bind_agent_to_core: True
 | 
						|
 | 
						|
wandb:
 | 
						|
  project: '{config[name]}'
 | 
						|
  sync_tensorboard: True
 | 
						|
  monitor_gym: True
 | 
						|
  save_code: False
 | 
						|
 | 
						|
video:
 | 
						|
  enable: True
 | 
						|
  length: 3000
 | 
						|
  frequency: 100
 | 
						|
 | 
						|
test:
 | 
						|
  enable: False
 | 
						|
  length: 3000
 | 
						|
  frequency: 100 # 32 # 10
 | 
						|
  deterministic: Both
 | 
						|
  num_envs: 1
 | 
						|
 | 
						|
env:
 | 
						|
  name: BoxPushingDense-v0
 | 
						|
  legacy_fancy: True
 | 
						|
  normalize_obs: True
 | 
						|
  normalize_rew: True
 | 
						|
  num_envs: 1
 | 
						|
  env_args:
 | 
						|
    more_obs: True
 | 
						|
 | 
						|
algo:
 | 
						|
  name: PPO
 | 
						|
  policy_name: MlpPolicy
 | 
						|
  n_steps: 4096
 | 
						|
  vf_coef: 1.0e-5
 | 
						|
  learning_rate: 5.0e-5
 | 
						|
  batch_size: 512
 | 
						|
  action_coef: 0
 | 
						|
  ent_coef: 0
 | 
						|
  normalize_advantage: False # True
 | 
						|
 | 
						|
pca:
 | 
						|
  enable: False
 | 
						|
  window: 64
 | 
						|
  skip_conditioning: True
 | 
						|
  Base_Noise: WHITE
 | 
						|
  init_std: 1.0
 | 
						|
 | 
						|
sweep:
 | 
						|
  enable: False
 | 
						|
  method: random
 | 
						|
  #metric:
 | 
						|
  #  goal: minimize
 | 
						|
  #  name: score
 | 
						|
  parameters:
 | 
						|
    algo.learning_rate:
 | 
						|
      min: 0.0001
 | 
						|
      max: 0.1
 | 
						|
---
 | 
						|
name: sweep
 | 
						|
import: $
 | 
						|
sweep.enable: True
 | 
						|
scheduler.reps_per_agent: 3
 | 
						|
---
 | 
						|
name: McNamo
 | 
						|
import: :DEFAULT
 | 
						|
 | 
						|
video:
 | 
						|
  length: 10
 | 
						|
 | 
						|
env:
 | 
						|
  env_args:
 | 
						|
    more_obs: False
 | 
						|
 | 
						|
algo.name: TRPL
 | 
						|
leaf: False
 | 
						|
---
 | 
						|
name: Leaf
 | 
						|
 | 
						|
vars:
 | 
						|
  leaf: True
 | 
						|
---
 | 
						|
name: Weird
 | 
						|
import: :McNamo,:Leaf
 | 
						|
 | 
						|
leaf: True
 | 
						|
---
 | 
						|
name: Test_BaseTest
 | 
						|
project: JustTesting
 | 
						|
runner: printConfig
 | 
						|
wandb:
 | 
						|
    group: 'base'
 | 
						|
    job_type: '{delta_desc}'
 | 
						|
---
 | 
						|
name: Test_Sweep
 | 
						|
project: JustTesting
 | 
						|
runner: printConfig
 | 
						|
scheduler:
 | 
						|
  reps_per_version: 1
 | 
						|
wandb:
 | 
						|
    group: 'sweep'
 | 
						|
    job_type: '{delta_desc}'
 | 
						|
A: 'dA'
 | 
						|
B: 'dB'
 | 
						|
C: 'dC'
 | 
						|
grid:
 | 
						|
  A: ['gA1', 'gA2']
 | 
						|
  # B: ['gB1', 'gB2']
 | 
						|
ablative:
 | 
						|
  C: ['aC1', 'aC2']
 | 
						|
---
 | 
						|
name: Test_CTR
 | 
						|
project: JustTesting
 | 
						|
runner: printConfig
 | 
						|
scheduler:
 | 
						|
  reps_per_version: 1
 | 
						|
  reps_per_agent: 2
 | 
						|
  agents_per_job: 2
 | 
						|
wandb:
 | 
						|
    group: 'ctr'
 | 
						|
    job_type: '{delta_desc}'
 | 
						|
CTR: 0
 | 
						|
ablative:
 | 
						|
  CTR: [1,2,3,4,5,6,7,8,9,10] |