slm_lab/spec/benchmark/a2c/a2c_nstep_cont.json
{
"a2c_nstep_bipedalwalker": {
"agent": [{
"name": "A2C",
"algorithm": {
"name": "ActorCritic",
"action_pdtype": "default",
"action_policy": "default",
"explore_var_spec": null,
"gamma": 0.99,
"lam": null,
"num_step_returns": 5,
"entropy_coef_spec": {
"name": "no_decay",
"start_val": 0.01,
"end_val": 0.01,
"start_step": 0,
"end_step": 0
},
"val_loss_coef": 0.5,
"training_frequency": 256
},
"memory": {
"name": "OnPolicyBatchReplay",
},
"net": {
"type": "MLPNet",
"shared": false,
"hid_layers": [256, 128],
"hid_layers_activation": "relu",
"init_fn": "orthogonal_",
"normalize": true,
"batch_norm": false,
"clip_grad_val": 0.5,
"use_same_optim": false,
"loss_spec": {
"name": "MSELoss"
},
"actor_optim_spec": {
"name": "Adam",
"lr": 3e-4,
},
"critic_optim_spec": {
"name": "Adam",
"lr": 3e-4,
},
"lr_scheduler_spec": null,
"gpu": false
}
}],
"env": [{
"name": "BipedalWalker-v2",
"num_envs": 32,
"max_t": null,
"max_frame": 4e6
}],
"body": {
"product": "outer",
"num": 1
},
"meta": {
"distributed": false,
"log_frequency": 10000,
"eval_frequency": 10000,
"max_session": 4,
"max_trial": 1
}
},
"a2c_nstep_pendulum": {
"agent": [{
"name": "A2C",
"algorithm": {
"name": "ActorCritic",
"action_pdtype": "default",
"action_policy": "default",
"explore_var_spec": null,
"gamma": 0.99,
"lam": null,
"num_step_returns": 5,
"entropy_coef_spec": {
"name": "no_decay",
"start_val": 0.0,
"end_val": 0.0,
"start_step": 0,
"end_step": 0
},
"val_loss_coef": 0.5,
"training_frequency": 2048
},
"memory": {
"name": "OnPolicyBatchReplay",
},
"net": {
"type": "MLPNet",
"shared": false,
"hid_layers": [64, 64],
"hid_layers_activation": "tanh",
"init_fn": "orthogonal_",
"normalize": false,
"batch_norm": false,
"clip_grad_val": 0.5,
"use_same_optim": false,
"loss_spec": {
"name": "MSELoss"
},
"actor_optim_spec": {
"name": "Adam",
"lr": 3e-4,
},
"critic_optim_spec": {
"name": "Adam",
"lr": 3e-4,
},
"lr_scheduler_spec": {
"name": "LinearToZero",
"frame": 1e6
},
"gpu": true
}
}],
"env": [{
"name": "Pendulum-v0",
"num_envs": 8,
"max_t": null,
"max_frame": 1e6
}],
"body": {
"product": "outer",
"num": 1
},
"meta": {
"distributed": false,
"log_frequency": 20000,
"eval_frequency": 20000,
"max_session": 4,
"max_trial": 1
}
}
}