config_ia2c_fp_net.ini

[MODEL_CONFIG]
rmsp_alpha = 0.99
rmsp_epsilon = 1e-5
max_grad_norm = 40
gamma = 0.99
lr_init = 5e-4
lr_decay = constant
entropy_coef = 0.01
value_coef = 0.5
num_lstm = 64
num_fc = 64
batch_size = 120
reward_norm = 100.0
reward_clip = -1

[TRAIN_CONFIG]
total_step = 1e6
test_interval = 2e6
log_interval = 1e4

[ENV_CONFIG]
clip_wave = -1
clip_wait = -1
control_interval_sec = 5
; agent is greedy, ia2c, ia2c_fp, ma2c_som, ma2c_ic3, ma2c_nc.
agent = ia2c_fp
; coop discount is used to discount the neighbors' impact
coop_gamma = 0.9
data_path = ./envs/real_net_data/
episode_length_sec = 3600
; for realnet, the normailization is done per agent
norm_wave = 1.0
norm_wait = -1
coef_wait = 0
flow_rate = 325
; objective is chosen from queue, wait, hybrid
objective = queue
scenario = atsc_real_net
seed = 12
test_seeds = 10000,20000,30000,40000,50000,60000,70000,80000,90000,100000
yellow_interval_sec = 2