diff --git a/cfg/rad_stoch.cfg b/cfg/rad_stoch.cfg index d8f12f6..eb7eec7 100644 --- a/cfg/rad_stoch.cfg +++ b/cfg/rad_stoch.cfg @@ -33,7 +33,6 @@ header = k, comm_radius, reward - [2, 1.25] k = 2 comm_radius = 1.25 diff --git a/cfg/transfer_stoch.cfg b/cfg/transfer_stoch.cfg index aed9061..bfdf9f6 100644 --- a/cfg/transfer_stoch.cfg +++ b/cfg/transfer_stoch.cfg @@ -1,9 +1,8 @@ [DEFAULT] - alg = dagger # learning parameters -batch_size = 40 +batch_size = 30 buffer_size = 10000 updates_per_step = 200 seed = 11 @@ -12,7 +11,7 @@ actor_lr = 5e-5 n_train_episodes = 400 beta_coeff = 0.993 test_interval = 40 -n_test_episodes = 40 +n_test_episodes = 20 # architecture parameters k = 3 @@ -28,8 +27,12 @@ n_agents = 50 n_actions = 2 n_states = 6 debug = True -header = reward +header = k, reward + +[4] +k = 4 +fname = transfer_stoch4 [3] k = 3