slm_lab/spec/experimental/reinforce/reinforce_cartpole.json
{
"reinforce_cartpole": {
"agent": [{
"name": "Reinforce",
"algorithm": {
"name": "Reinforce",
"action_pdtype": "default",
"action_policy": "default",
"explore_var_spec": null,
"gamma": 0.99,
"entropy_coef_spec": {
"name": "linear_decay",
"start_val": 0.01,
"end_val": 0.001,
"start_step": 0,
"end_step": 20000,
},
"training_frequency": 1
},
"memory": {
"name": "OnPolicyReplay"
},
"net": {
"type": "MLPNet",
"hid_layers": [64],
"hid_layers_activation": "selu",
"clip_grad_val": null,
"loss_spec": {
"name": "MSELoss"
},
"optim_spec": {
"name": "Adam",
"lr": 0.002
},
"lr_scheduler_spec": null
}
}],
"env": [{
"name": "CartPole-v0",
"max_t": null,
"max_frame": 80000,
}],
"body": {
"product": "outer",
"num": 1
},
"meta": {
"distributed": false,
"eval_frequency": 2000,
"max_session": 1,
"max_trial": 1,
}
},
"reinforce_cross_entropy_cartpole": {
"agent": [{
"name": "Reinforce",
"algorithm": {
"name": "Reinforce",
"action_pdtype": "Categorical",
"action_policy": "default",
"center_return": true,
"explore_var_spec": null,
"gamma": 0.99,
"entropy_coef_spec": {
"name": "linear_decay",
"start_val": 0.01,
"end_val": 0.001,
"start_step": 0,
"end_step": 20000,
},
"training_frequency": 16
},
"memory": {
"name": "OnPolicyCrossEntropy",
"cross_entropy": 0.5
},
"net": {
"type": "MLPNet",
"hid_layers": [128],
"hid_layers_activation": "selu",
"clip_grad_val": null,
"loss_spec": {
"name": "CrossEntropyLoss"
},
"optim_spec": {
"name": "Adam",
"lr": 0.02
},
"lr_scheduler_spec": null
}
}],
"env": [{
"name": "CartPole-v0",
"max_t": null,
"max_frame": 100000,
}],
"body": {
"product": "outer",
"num": 1
},
"meta": {
"distributed": false,
"eval_frequency": 2000,
"max_session": 4,
"max_trial": 1,
},
"search": {
"agent": [{
"algorithm": {
"gamma__grid_search": [0.1, 0.5, 0.7, 0.8, 0.90, 0.99, 0.999]
}
}]
}
}
}