johnjim0816
add all benchmarks
62e03a2
raw
history blame
836 Bytes
general_cfg:
algo_name: DQN
collect_traj: true
device: cpu
env_name: gym
load_checkpoint: true
load_model_step: best
load_path: Train_single_CartPole-v1_DQN_20230515-211721
max_episode: 10
max_step: 200
mode: test
model_save_fre: 500
mp_backend: single
n_workers: 1
online_eval: true
online_eval_episode: 10
save_fig: true
seed: 1
show_fig: false
algo_cfg:
batch_size: 64
buffer_size: 100000
buffer_type: REPLAY_QUE
epsilon_decay: 500
epsilon_end: 0.01
epsilon_start: 0.95
gamma: 0.95
lr: 0.0001
target_update: 4
value_layers:
- activation: relu
layer_dim:
- 256
layer_type: linear
- activation: relu
layer_dim:
- 256
layer_type: linear
env_cfg:
id: CartPole-v1
ignore_params:
- wrapper
- ignore_params
render_mode: rgb_array
wrapper: null