diff --git a/configs/config.py b/configs/config.py
index 0b0021d5b4b249b390c4000e926fc3ed7af4b66d..16b2ff802e5be18f8307380ce8dedfde19d8e867 100644
--- a/configs/config.py
+++ b/configs/config.py
@@ -301,7 +301,7 @@ def get_config():
parser.add_argument("--log_interval", type=int,
default=1, help="time duration between contiunous twice log printing.")
parser.add_argument("--model_dir", type=str,
- default="./trained_models/7749_threenights/run1/models", help="by default None. set the path to pretrained model.")
+ default="./trained_models/9191_fournights/run1/models", help="by default None. set the path to pretrained model.")
# eval parameters
parser.add_argument("--use_eval", action='store_true',
diff --git a/trained_models/9191_fournights/config.py b/trained_models/9191_fournights/config.py
new file mode 100644
index 0000000000000000000000000000000000000000..2d40fcd244a00fa0713f8da2e9b6a5468d7b3559
--- /dev/null
+++ b/trained_models/9191_fournights/config.py
@@ -0,0 +1,330 @@
+import argparse
+from behaviour import behaviour_choices
+
+def get_config():
+ """
+ The configuration parser for common hyperparameters of all environment.
+ Please reach each `scripts/train/<env>_runner.py` file to find private hyperparameters
+ only used in <env>.
+
+ Prepare parameters:
+ --algorithm_name <algorithm_name>
+ specifiy the algorithm, including `["happo", "hatrpo"]`
+ --experiment_name <str>
+ an identifier to distinguish different experiment.
+ --seed <int>
+ set seed for numpy and torch
+ --seed_specify
+ by default True Random or specify seed for numpy/torch
+ --runing_id <int>
+ the runing index of experiment (default=1)
+ --cuda
+ by default True, will use GPU to train; or else will use CPU;
+ --cuda_deterministic
+ by default, make sure random seed effective. if set, bypass such function.
+ --n_training_threads <int>
+ number of training threads working in parallel. by default 1
+ --n_rollout_threads <int>
+ number of parallel envs for training rollout. by default 32
+ --n_eval_rollout_threads <int>
+ number of parallel envs for evaluating rollout. by default 1
+ --n_render_rollout_threads <int>
+ number of parallel envs for rendering, could only be set as 1 for some environments.
+ --num_env_steps <int>
+ number of env steps to train (default: 10e6)
+
+
+ Env parameters:
+ --env_name <str>
+ specify the name of environment
+ --use_obs_instead_of_state
+ [only for some env] by default False, will use global state; or else will use concatenated local obs.
+
+ Replay Buffer parameters:
+ --episode_length <int>
+ the max length of episode in the buffer.
+
+ Network parameters:
+ --share_policy
+ by default True, all agents will share the same network; set to make training agents use different policies.
+ --use_centralized_V
+ by default True, use centralized training mode; or else will decentralized training mode.
+ --stacked_frames <int>
+ Number of input frames which should be stack together.
+ --hidden_size <int>
+ Dimension of hidden layers for actor/critic networks
+ --layer_N <int>
+ Number of layers for actor/critic networks
+ --use_ReLU
+ by default True, will use ReLU. or else will use Tanh.
+ --use_popart
+ by default True, use running mean and std to normalize rewards.
+ --use_feature_normalization
+ by default True, apply layernorm to normalize inputs.
+ --use_orthogonal
+ by default True, use Orthogonal initialization for weights and 0 initialization for biases. or else, will use xavier uniform inilialization.
+ --gain
+ by default 0.01, use the gain # of last action layer
+ --use_naive_recurrent_policy
+ by default False, use the whole trajectory to calculate hidden states.
+ --use_recurrent_policy
+ by default, use Recurrent Policy. If set, do not use.
+ --recurrent_N <int>
+ The number of recurrent layers ( default 1).
+ --data_chunk_length <int>
+ Time length of chunks used to train a recurrent_policy, default 10.
+
+ Optimizer parameters:
+ --lr <float>
+ learning rate parameter, (default: 5e-4, fixed).
+ --critic_lr <float>
+ learning rate of critic (default: 5e-4, fixed)
+ --opti_eps <float>
+ RMSprop optimizer epsilon (default: 1e-5)
+ --weight_decay <float>
+ coefficience of weight decay (default: 0)
+
+ TRPO parameters:
+ --kl_threshold <float>
+ the threshold of kl-divergence (default: 0.01)
+ --ls_step <int>
+ the step of line search (default: 10)
+ --accept_ratio <float>
+ accept ratio of loss improve (default: 0.5)
+
+ PPO parameters:
+ --ppo_epoch <int>
+ number of ppo epochs (default: 15)
+ --use_clipped_value_loss
+ by default, clip loss value. If set, do not clip loss value.
+ --clip_param <float>
+ ppo clip parameter (default: 0.2)
+ --num_mini_batch <int>
+ number of batches for ppo (default: 1)
+ --entropy_coef <float>
+ entropy term coefficient (default: 0.01)
+ --use_max_grad_norm
+ by default, use max norm of gradients. If set, do not use.
+ --max_grad_norm <float>
+ max norm of gradients (default: 0.5)
+ --use_gae
+ by default, use generalized advantage estimation. If set, do not use gae.
+ --gamma <float>
+ discount factor for rewards (default: 0.99)
+ --gae_lambda <float>
+ gae lambda parameter (default: 0.95)
+ --use_proper_time_limits
+ by default, the return value does consider limits of time. If set, compute returns with considering time limits factor.
+ --use_huber_loss
+ by default, use huber loss. If set, do not use huber loss.
+ --use_value_active_masks
+ by default True, whether to mask useless data in value loss.
+ --huber_delta <float>
+ coefficient of huber loss.
+
+
+ Run parameters：
+ --use_linear_lr_decay
+ by default, do not apply linear decay to learning rate. If set, use a linear schedule on the learning rate
+ --save_interval <int>
+ time duration between contiunous twice models saving.
+ --log_interval <int>
+ time duration between contiunous twice log printing.
+ --model_dir <str>
+ by default None. set the path to pretrained model.
+
+ Eval parameters:
+ --use_eval
+ by default, do not start evaluation. If set`, start evaluation alongside with training.
+ --eval_interval <int>
+ time duration between contiunous twice evaluation progress.
+ --eval_episodes <int>
+ number of episodes of a single evaluation.
+
+ Render parameters:
+ --save_gifs
+ by default, do not save render video. If set, save video.
+ --use_render
+ by default, do not render the env during training. If set, start render. Note: something, the environment has internal render process which is not controlled by this hyperparam.
+ --render_episodes <int>
+ the number of episodes to render a given env
+ --ifi <float>
+ the play interval of each rendered image in saved video.
+
+ Pretrained parameters:
+
+ """
+ parser = argparse.ArgumentParser(description='onpolicy_algorithm', formatter_class=argparse.RawDescriptionHelpFormatter)
+
+ # prepare parameters
+ parser.add_argument("--algorithm_name", type=str,
+ default='happo', choices=["happo","hatrpo"])
+ parser.add_argument("--experiment_name", type=str,
+ default="check", help="an identifier to distinguish different experiment.")
+ parser.add_argument("--seed", type=int,
+ default=1, help="Random seed for numpy/torch")
+ parser.add_argument("--seed_specify", action="store_true",
+ default=False, help="Random or specify seed for numpy/torch")
+ parser.add_argument("--red_behaviour", choices=behaviour_choices, type=str.lower,
+ default="simpleone", help="name of red behaviour to use for training. Ignored if rand_red_behaviour is True.")
+ parser.add_argument("--rand_red_behaviour", type=bool,
+ default=True, help="Randomizes opponent behaviour in each simulation for training. Uses red_behaviour if False.")
+ parser.add_argument("--rand_size", type=bool,
+ default=True, help="Randomizes both teams, teamsize in each simulation for training.")
+ parser.add_argument("--red_size", type=int,
+ default=7, help="2<=Size of red team<=7, works only with rand_size==false")
+ parser.add_argument("--green_size", type=int,
+ default=7, help="2<=Size of green team<=7, works only with rand_size==false")
+ parser.add_argument("--runing_id", type=int,
+ default=1, help="the runing index of experiment")
+ parser.add_argument("--cuda", action='store_false',
+ default=True, help="by default True, will use GPU to train; or else will use CPU;")
+ parser.add_argument("--cuda_deterministic", action='store_false',
+ default=True, help="by default, make sure random seed effective. if set, bypass such function.")
+ parser.add_argument("--n_training_threads", type=int,
+ default=20, help="Number of torch threads for training")
+ parser.add_argument("--n_rollout_threads", type=int,
+ default=1, help="Number of parallel envs for training rollouts")
+ parser.add_argument("--n_eval_rollout_threads", type=int,
+ default=1, help="Number of parallel envs for evaluating rollouts")
+ parser.add_argument("--n_render_rollout_threads", type=int,
+ default=1, help="Number of parallel envs for rendering rollouts")
+ parser.add_argument("--num_env_steps", type=int,
+ default=60000000, help='Number of environment steps to train (default: 6000)')
+ parser.add_argument("--user_name", type=str,
+ default='marl',help="[for wandb usage], to specify user's name for simply collecting training data.")
+ # env parameters
+ parser.add_argument("--env_name", type=str,
+ default='scontrol', help="specify the name of environment")
+ parser.add_argument("--use_obs_instead_of_state", action='store_true',
+ default=False, help="Whether to use global state or concatenated obs")
+
+ # replay buffer parameters
+ parser.add_argument("--episode_length", type=int,
+ default=600, help="Max length for any episode")
+
+ # network parameters
+ parser.add_argument("--share_policy", action='store_false',
+ default=True, help='Whether agent share the same policy')
+ parser.add_argument("--use_centralized_V", action='store_false',
+ default=True, help="Whether to use centralized V function")
+ parser.add_argument("--stacked_frames", type=int,
+ default=100, help="Dimension of hidden layers for actor/critic networks")
+ parser.add_argument("--use_stacked_frames", action='store_true',
+ default=True, help="Whether to use stacked_frames")
+ parser.add_argument("--hidden_size", type=int,
+ default=64, help="Dimension of hidden layers for actor/critic networks")
+ parser.add_argument("--layer_N", type=int,
+ default=2, help="Number of layers for actor/critic networks")
+ parser.add_argument("--use_ReLU", action='store_false',
+ default=True, help="Whether to use ReLU")
+ parser.add_argument("--use_popart", action='store_false',
+ default=True, help="by default True, use running mean and std to normalize rewards.")
+ parser.add_argument("--use_valuenorm", action='store_false',
+ default=True, help="by default True, use running mean and std to normalize rewards.")
+ parser.add_argument("--use_feature_normalization", action='store_false',
+ default=True, help="Whether to apply layernorm to the inputs")
+ parser.add_argument("--use_orthogonal", action='store_false',
+ default=True, help="Whether to use Orthogonal initialization for weights and 0 initialization for biases")
+ parser.add_argument("--gain", type=float,
+ default=0.01, help="The gain # of last action layer")
+
+ # recurrent parameters
+ parser.add_argument("--use_naive_recurrent_policy", action='store_true',
+ default=False, help='Whether to use a naive recurrent policy')
+ parser.add_argument("--use_recurrent_policy", action='store_true',
+ default=False, help='use a recurrent policy')
+ parser.add_argument("--recurrent_N", type=int,
+ default=1, help="The number of recurrent layers.")
+ parser.add_argument("--data_chunk_length", type=int,
+ default=10, help="Time length of chunks used to train a recurrent_policy")
+
+ # optimizer parameters
+ parser.add_argument("--lr", type=float,
+ default=1e-4, help='learning rate (default: 5e-4)')
+ parser.add_argument("--critic_lr", type=float,
+ default=1e-4, help='critic learning rate (default: 5e-4)')
+ parser.add_argument("--opti_eps", type=float,
+ default=1e-5, help='RMSprop optimizer epsilon (default: 1e-5)')
+ parser.add_argument("--weight_decay", type=float, default=0)
+ parser.add_argument("--std_x_coef", type=float, default=1)
+ parser.add_argument("--std_y_coef", type=float, default=0.5)
+
+
+ # trpo parameters
+ parser.add_argument("--kl_threshold", type=float,
+ default=0.01, help='the threshold of kl-divergence (default: 0.01)')
+ parser.add_argument("--ls_step", type=int,
+ default=10, help='number of line search (default: 10)')
+ parser.add_argument("--accept_ratio", type=float,
+ default=0.5, help='accept ratio of loss improve (default: 0.5)')
+
+ # ppo parameters
+ parser.add_argument("--ppo_epoch", type=int,
+ default=15, help='number of ppo epochs (default: 15)')
+ parser.add_argument("--use_clipped_value_loss", action='store_false',
+ default=True, help="by default, clip loss value. If set, do not clip loss value.")
+ parser.add_argument("--clip_param", type=float,
+ default=0.2, help='ppo clip parameter (default: 0.2)')
+ parser.add_argument("--num_mini_batch", type=int,
+ default=1, help='number of batches for ppo (default: 1)')
+ parser.add_argument("--entropy_coef", type=float,
+ default=0.01, help='entropy term coefficient (default: 0.01)')
+ parser.add_argument("--value_loss_coef", type=float,
+ default=0.01, help='value loss coefficient (default: 0.5)')
+ parser.add_argument("--use_max_grad_norm", action='store_false',
+ default=True, help="by default, use max norm of gradients. If set, do not use.")
+ parser.add_argument("--max_grad_norm", type=float,
+ default=10.0, help='max norm of gradients (default: 0.5)')
+ parser.add_argument("--use_gae", action='store_false',
+ default=True, help='use generalized advantage estimation')
+ parser.add_argument("--gamma", type=float, default=0.999999999,
+ help='discount factor for rewards (default: 0.99)')
+ parser.add_argument("--gae_lambda", type=float, default=0.95,
+ help='gae lambda parameter (default: 0.95)')
+ parser.add_argument("--use_proper_time_limits", action='store_true',
+ default=False, help='compute returns taking into account time limits')
+ parser.add_argument("--use_huber_loss", action='store_false',
+ default=True, help="by default, use huber loss. If set, do not use huber loss.")
+ parser.add_argument("--use_value_active_masks", action='store_false',
+ default=True, help="by default True, whether to mask useless data in value loss.")
+ parser.add_argument("--use_policy_active_masks", action='store_false',
+ default=True, help="by default True, whether to mask useless data in policy loss.")
+ parser.add_argument("--huber_delta", type=float,
+ default=10.0, help=" coefficience of huber loss.")
+
+ # run parameters
+ parser.add_argument("--use_linear_lr_decay", action='store_true',
+ default=False, help='use a linear schedule on the learning rate')
+ parser.add_argument("--save_interval", type=int,
+ default=1, help="time duration between contiunous twice models saving.")
+ parser.add_argument("--log_interval", type=int,
+ default=1, help="time duration between contiunous twice log printing.")
+ parser.add_argument("--model_dir", type=str,
+ default="./trained_models/7749_threenights/run1/models", help="by default None. set the path to pretrained model.")
+
+ # eval parameters
+ parser.add_argument("--use_eval", action='store_true',
+ default=True, help="by default, do not start evaluation. If set`, start evaluation alongside with training.")
+ parser.add_argument("--eval_interval", type=int,
+ default=10, help="time duration between contiunous twice evaluation progress.")
+ parser.add_argument("--eval_episodes", type=int,
+ default=2, help="number of episodes of a single evaluation.")
+ parser.add_argument("--eval_render", type=bool,
+ default=False, help="visualizes the model every eval. works on top of the 'eval' setting in multiagent_rl_train.")
+ parser.add_argument("--visualise_delay", type=int,
+ default=1, help="numer of milliseconds to wait between steps when drawing eval render.")
+ parser.add_argument("--only_delay_middle", type=bool,
+ default=True, help="uses 0 delay for time < 150 and time > 350, and visualise delay for the middle")
+
+ # render parameters
+ parser.add_argument("--save_gifs", action='store_true',
+ default=False, help="by default, do not save render video. If set, save video.")
+ parser.add_argument("--use_render", action='store_true',
+ default=False, help="by default, do not render the env during training. If set, start render. Note: something, the environment has internal render process which is not controlled by this hyperparam.")
+ parser.add_argument("--render_episodes", type=int,
+ default=5, help="the number of episodes to render a given env")
+ parser.add_argument("--ifi", type=float,
+ default=0.1, help="the play interval of each rendered image in saved video.")
+
+ return parser
\ No newline at end of file
diff --git a/trained_models/9191_fournights/reward b/trained_models/9191_fournights/reward
new file mode 100644
index 0000000000000000000000000000000000000000..de5eaace25ee018032d05f719a55edb6c529277b
--- /dev/null
+++ b/trained_models/9191_fournights/reward
@@ -0,0 +1,10 @@
+agent_id: + reward[agent_id]['share_radar_actions'] + reward[agent_id]['detected_with_no_radar']
+ + reward[agent_id]['detected_multiple']
+ + (0.01*reward[agent_id]['position_error'] - 0.01*reward[agent_id]['enemy_position_error'])
+eval: agent_id: (0.01*reward[agent_id]['position_error'] - 0.01*reward[agent_id]['enemy_position_error'])
+ + (0.01*reward[agent_id]['position_delay_error'] - 0.01*reward[agent_id]['enemy_position_delay_error'])
+ + (reward[agent_id]['aircraft_share_detected']-reward[agent_id]['enemy_aircraft_share_detected'])
+ + (reward[agent_id]['aircraft_share_delay_detected']-reward[agent_id]['enemy_aircraft_share_delay_detected'])
+ + (0.1*reward[agent_id]['radar_share_detected']-0.1*reward[agent_id]['enemy_radar_share_detected'])
+ + (0.1*reward[agent_id]['radar_share_delay_detected']-0.1*reward[agent_id]['enemy_radar_share_delay_detected'])
+ + (reward[agent_id]['share_radar_actions'] - reward[agent_id]['enemy_share_radar_actions'])
diff --git a/trained_models/9191_fournights/run1/logs/agent0/actor_grad_norm/agent0/actor_grad_norm/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent0/actor_grad_norm/agent0/actor_grad_norm/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..f182b9800671a6af14ac9641a1736d27d880d149
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent0/actor_grad_norm/agent0/actor_grad_norm/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent0/average_step_adversarial_rewards/agent0/average_step_adversarial_rewards/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent0/average_step_adversarial_rewards/agent0/average_step_adversarial_rewards/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..8308e715defcc6fe9248a16a9ae7b111732cb653
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent0/average_step_adversarial_rewards/agent0/average_step_adversarial_rewards/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent0/average_step_rewards/agent0/average_step_rewards/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent0/average_step_rewards/agent0/average_step_rewards/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..4984f068655e5003dd6c0c42056ac10ebe23dd39
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent0/average_step_rewards/agent0/average_step_rewards/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent0/critic_grad_norm/agent0/critic_grad_norm/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent0/critic_grad_norm/agent0/critic_grad_norm/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..1b8cfdc7c5b088a778e920063bfa726ac5aac21c
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent0/critic_grad_norm/agent0/critic_grad_norm/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent0/dist_entropy/agent0/dist_entropy/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent0/dist_entropy/agent0/dist_entropy/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..285510669932610dcba85ebc5cc8e9223ea0d223
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent0/dist_entropy/agent0/dist_entropy/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent0/policy_loss/agent0/policy_loss/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent0/policy_loss/agent0/policy_loss/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..e9efb1c6057d8c24b25ab0534ffd9a25196eeed2
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent0/policy_loss/agent0/policy_loss/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent0/ratio/agent0/ratio/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent0/ratio/agent0/ratio/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..4c8ccb0f052f3a89f46bbe5c9f6f19a4eb18ae77
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent0/ratio/agent0/ratio/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent0/value_loss/agent0/value_loss/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent0/value_loss/agent0/value_loss/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..3f4a528430c21958963bd647f4b91b0bcc713398
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent0/value_loss/agent0/value_loss/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent1/actor_grad_norm/agent1/actor_grad_norm/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent1/actor_grad_norm/agent1/actor_grad_norm/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..4986f8a31e5f5b56a229101b3a87fcd6fa2ce20f
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent1/actor_grad_norm/agent1/actor_grad_norm/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent1/average_step_adversarial_rewards/agent1/average_step_adversarial_rewards/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent1/average_step_adversarial_rewards/agent1/average_step_adversarial_rewards/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..8ea299d81cd36ea43597bee36d7431e21907ce43
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent1/average_step_adversarial_rewards/agent1/average_step_adversarial_rewards/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent1/average_step_rewards/agent1/average_step_rewards/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent1/average_step_rewards/agent1/average_step_rewards/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..b5c1a03f908a680f139e6017d5706af31de0eab6
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent1/average_step_rewards/agent1/average_step_rewards/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent1/critic_grad_norm/agent1/critic_grad_norm/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent1/critic_grad_norm/agent1/critic_grad_norm/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..0f61431aacc523aae8810bd883ff89c4992518d7
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent1/critic_grad_norm/agent1/critic_grad_norm/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent1/dist_entropy/agent1/dist_entropy/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent1/dist_entropy/agent1/dist_entropy/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..00a09d7e7d10715d2e88e12ab7f71e22d5796546
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent1/dist_entropy/agent1/dist_entropy/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent1/policy_loss/agent1/policy_loss/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent1/policy_loss/agent1/policy_loss/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..d8ed8324c620a51315d85699a595e3e2176b2a4f
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent1/policy_loss/agent1/policy_loss/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent1/ratio/agent1/ratio/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent1/ratio/agent1/ratio/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..a1088b55846d5c6aee14b5a23991e65beed2da46
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent1/ratio/agent1/ratio/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent1/value_loss/agent1/value_loss/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent1/value_loss/agent1/value_loss/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..2a802386f21de51dddef7d35f8605c91ac444eb9
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent1/value_loss/agent1/value_loss/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent2/actor_grad_norm/agent2/actor_grad_norm/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent2/actor_grad_norm/agent2/actor_grad_norm/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..75b08ebeabb1a48b0b1df2d02d23499ff9bb52d0
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent2/actor_grad_norm/agent2/actor_grad_norm/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent2/average_step_adversarial_rewards/agent2/average_step_adversarial_rewards/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent2/average_step_adversarial_rewards/agent2/average_step_adversarial_rewards/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..073bbd6311d0c447775a673689551a1b3a2c8959
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent2/average_step_adversarial_rewards/agent2/average_step_adversarial_rewards/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent2/average_step_rewards/agent2/average_step_rewards/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent2/average_step_rewards/agent2/average_step_rewards/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..36478ab39f048a4905b5b5badcaf7addb0fe02e7
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent2/average_step_rewards/agent2/average_step_rewards/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent2/critic_grad_norm/agent2/critic_grad_norm/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent2/critic_grad_norm/agent2/critic_grad_norm/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..6a9da58b85aefae77928b32f3e353e4625c123fa
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent2/critic_grad_norm/agent2/critic_grad_norm/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent2/dist_entropy/agent2/dist_entropy/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent2/dist_entropy/agent2/dist_entropy/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..02ed869294e1329d3574784d85a10e7e389ed091
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent2/dist_entropy/agent2/dist_entropy/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent2/policy_loss/agent2/policy_loss/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent2/policy_loss/agent2/policy_loss/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..cb6e9ee4725eda6e2645598b68b9c41b6687e6d1
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent2/policy_loss/agent2/policy_loss/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent2/ratio/agent2/ratio/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent2/ratio/agent2/ratio/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..fd0a76b0efbf25f1de3610be17c4ce49d41bee73
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent2/ratio/agent2/ratio/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent2/value_loss/agent2/value_loss/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent2/value_loss/agent2/value_loss/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..1c40c48451574d987e6292f083b6ad34631068e1
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent2/value_loss/agent2/value_loss/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent3/actor_grad_norm/agent3/actor_grad_norm/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent3/actor_grad_norm/agent3/actor_grad_norm/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..9481b488861b8141162b314a0e8d5eee6c1a30f3
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent3/actor_grad_norm/agent3/actor_grad_norm/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent3/average_step_adversarial_rewards/agent3/average_step_adversarial_rewards/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent3/average_step_adversarial_rewards/agent3/average_step_adversarial_rewards/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..8ff364a254f6034030814016b752946787a5cb10
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent3/average_step_adversarial_rewards/agent3/average_step_adversarial_rewards/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent3/average_step_rewards/agent3/average_step_rewards/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent3/average_step_rewards/agent3/average_step_rewards/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..76486bb2d2e1ead3bfad640975e362605b8562f8
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent3/average_step_rewards/agent3/average_step_rewards/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent3/critic_grad_norm/agent3/critic_grad_norm/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent3/critic_grad_norm/agent3/critic_grad_norm/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..592fd9831c3a920d20b96e511ba1dc6eab2c5aa5
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent3/critic_grad_norm/agent3/critic_grad_norm/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent3/dist_entropy/agent3/dist_entropy/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent3/dist_entropy/agent3/dist_entropy/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..2a13ecb81df9474995382ed709da9213dc5e277a
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent3/dist_entropy/agent3/dist_entropy/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent3/policy_loss/agent3/policy_loss/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent3/policy_loss/agent3/policy_loss/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..105c3588f2bafab8bea9ffd19d4d004be9a44323
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent3/policy_loss/agent3/policy_loss/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent3/ratio/agent3/ratio/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent3/ratio/agent3/ratio/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..e43d3952337ad355c7033de9aaff982105d7933f
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent3/ratio/agent3/ratio/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent3/value_loss/agent3/value_loss/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent3/value_loss/agent3/value_loss/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..eb3388dd6133077023486612d4e415136ed2df5d
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent3/value_loss/agent3/value_loss/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent4/actor_grad_norm/agent4/actor_grad_norm/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent4/actor_grad_norm/agent4/actor_grad_norm/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..ff4a658e64bd9092905314769e824f416e260461
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent4/actor_grad_norm/agent4/actor_grad_norm/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent4/average_step_adversarial_rewards/agent4/average_step_adversarial_rewards/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent4/average_step_adversarial_rewards/agent4/average_step_adversarial_rewards/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..af3aacd6c7a185e330c7b1e723b511c6ab89a4b6
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent4/average_step_adversarial_rewards/agent4/average_step_adversarial_rewards/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent4/average_step_rewards/agent4/average_step_rewards/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent4/average_step_rewards/agent4/average_step_rewards/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..a3552c00a642bff67950a8837615a239a2453432
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent4/average_step_rewards/agent4/average_step_rewards/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent4/critic_grad_norm/agent4/critic_grad_norm/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent4/critic_grad_norm/agent4/critic_grad_norm/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..beb9fc5c73ec5b7a0f102b019ed9e2ab9e76f7fa
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent4/critic_grad_norm/agent4/critic_grad_norm/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent4/dist_entropy/agent4/dist_entropy/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent4/dist_entropy/agent4/dist_entropy/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..5fedbc4bc44b8d179cac072b1f3ae21be591ae87
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent4/dist_entropy/agent4/dist_entropy/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent4/policy_loss/agent4/policy_loss/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent4/policy_loss/agent4/policy_loss/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..a30d01f82fed2a7fdd8aa6a0d673cd8c0a053e86
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent4/policy_loss/agent4/policy_loss/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent4/ratio/agent4/ratio/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent4/ratio/agent4/ratio/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..3032cfe3eafe03dabccdf5232f55a510f3a2a77f
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent4/ratio/agent4/ratio/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent4/value_loss/agent4/value_loss/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent4/value_loss/agent4/value_loss/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..6e2f731f2b59f579d92ad33eb4e7ddcf737194c6
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent4/value_loss/agent4/value_loss/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent5/actor_grad_norm/agent5/actor_grad_norm/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent5/actor_grad_norm/agent5/actor_grad_norm/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..61afc077cd4a684adcca6190d83a3f9793c86eb4
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent5/actor_grad_norm/agent5/actor_grad_norm/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent5/average_step_adversarial_rewards/agent5/average_step_adversarial_rewards/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent5/average_step_adversarial_rewards/agent5/average_step_adversarial_rewards/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..d8877007b7c770e35228ed8cefada6013c3a7d37
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent5/average_step_adversarial_rewards/agent5/average_step_adversarial_rewards/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent5/average_step_rewards/agent5/average_step_rewards/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent5/average_step_rewards/agent5/average_step_rewards/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..efec4cb41d8fe512c2f606186268263c08499976
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent5/average_step_rewards/agent5/average_step_rewards/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent5/critic_grad_norm/agent5/critic_grad_norm/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent5/critic_grad_norm/agent5/critic_grad_norm/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..586caaac818390f16a04e22f31e2d935a61d3b05
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent5/critic_grad_norm/agent5/critic_grad_norm/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent5/dist_entropy/agent5/dist_entropy/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent5/dist_entropy/agent5/dist_entropy/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..7f2088816448d58c548d6acc2f6a07949f00df82
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent5/dist_entropy/agent5/dist_entropy/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent5/policy_loss/agent5/policy_loss/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent5/policy_loss/agent5/policy_loss/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..de4a4d2c5a14c25185a1df38c5a8cf6632003057
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent5/policy_loss/agent5/policy_loss/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent5/ratio/agent5/ratio/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent5/ratio/agent5/ratio/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..efb67ce319e1469288c104e762dc924edffe008a
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent5/ratio/agent5/ratio/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent5/value_loss/agent5/value_loss/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent5/value_loss/agent5/value_loss/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..a9eb46c86004a90f88cf03cbd98d313cc0aca156
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent5/value_loss/agent5/value_loss/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent6/actor_grad_norm/agent6/actor_grad_norm/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent6/actor_grad_norm/agent6/actor_grad_norm/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..54a76d1a7a9268b5b32b35cfd9874242f5e465d3
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent6/actor_grad_norm/agent6/actor_grad_norm/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent6/average_step_adversarial_rewards/agent6/average_step_adversarial_rewards/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent6/average_step_adversarial_rewards/agent6/average_step_adversarial_rewards/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..56e9a9e0f3b66db48f82a045ec2cb0d9d078ea96
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent6/average_step_adversarial_rewards/agent6/average_step_adversarial_rewards/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent6/average_step_rewards/agent6/average_step_rewards/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent6/average_step_rewards/agent6/average_step_rewards/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..b72876fe3b67578e309014a673e07c534f73d776
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent6/average_step_rewards/agent6/average_step_rewards/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent6/critic_grad_norm/agent6/critic_grad_norm/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent6/critic_grad_norm/agent6/critic_grad_norm/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..36aad933e2ffd5ba32ff0c5df77c887b77930b97
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent6/critic_grad_norm/agent6/critic_grad_norm/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent6/dist_entropy/agent6/dist_entropy/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent6/dist_entropy/agent6/dist_entropy/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..3a2eaa9432be6bd8afe3c135d644d72699c152d5
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent6/dist_entropy/agent6/dist_entropy/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent6/policy_loss/agent6/policy_loss/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent6/policy_loss/agent6/policy_loss/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..2b87dc06ba4883b866dcca8113a3f5c20b521dc3
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent6/policy_loss/agent6/policy_loss/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent6/ratio/agent6/ratio/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent6/ratio/agent6/ratio/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..bbdef595e14ed2577e12fb6663e9bafd28ef2bbf
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent6/ratio/agent6/ratio/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/agent6/value_loss/agent6/value_loss/events.out.tfevents.1670950489.maiex1-104 b/trained_models/9191_fournights/run1/logs/agent6/value_loss/agent6/value_loss/events.out.tfevents.1670950489.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..657eb1df0dfdf34367e1fb18eb18b85b6e680800
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/agent6/value_loss/agent6/value_loss/events.out.tfevents.1670950489.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/eval_average_episode_rewards/eval_average_episode_rewards/events.out.tfevents.1670951582.maiex1-104 b/trained_models/9191_fournights/run1/logs/eval_average_episode_rewards/eval_average_episode_rewards/events.out.tfevents.1670951582.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..a213692cae75a40d95dd2cf0a6380868878b06e0
Binary files /dev/null and b/trained_models/9191_fournights/run1/logs/eval_average_episode_rewards/eval_average_episode_rewards/events.out.tfevents.1670951582.maiex1-104 differ
diff --git a/trained_models/9191_fournights/run1/logs/events.out.tfevents.1670950364.maiex1-104 b/trained_models/9191_fournights/run1/logs/events.out.tfevents.1670950364.maiex1-104
new file mode 100644
index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391
diff --git a/trained_models/9191_fournights/run1/models/actor_agent0.pt b/trained_models/9191_fournights/run1/models/actor_agent0.pt
new file mode 100644
index 0000000000000000000000000000000000000000..3b42cd00a4ef535912373d43a02fea2d00ae3ba6
Binary files /dev/null and b/trained_models/9191_fournights/run1/models/actor_agent0.pt differ
diff --git a/trained_models/9191_fournights/run1/models/actor_agent1.pt b/trained_models/9191_fournights/run1/models/actor_agent1.pt
new file mode 100644
index 0000000000000000000000000000000000000000..f0b7df31b604174c8a2f17d149eaee65ac6cae84
Binary files /dev/null and b/trained_models/9191_fournights/run1/models/actor_agent1.pt differ
diff --git a/trained_models/9191_fournights/run1/models/actor_agent2.pt b/trained_models/9191_fournights/run1/models/actor_agent2.pt
new file mode 100644
index 0000000000000000000000000000000000000000..6416145f54fa8d89292236a32c849cf4d9a437e0
Binary files /dev/null and b/trained_models/9191_fournights/run1/models/actor_agent2.pt differ
diff --git a/trained_models/9191_fournights/run1/models/actor_agent3.pt b/trained_models/9191_fournights/run1/models/actor_agent3.pt
new file mode 100644
index 0000000000000000000000000000000000000000..4fcefafe4c801dbf4c82fb919f8662d76211d0e5
Binary files /dev/null and b/trained_models/9191_fournights/run1/models/actor_agent3.pt differ
diff --git a/trained_models/9191_fournights/run1/models/actor_agent4.pt b/trained_models/9191_fournights/run1/models/actor_agent4.pt
new file mode 100644
index 0000000000000000000000000000000000000000..baa8d53e803306fbb651f8f53f529e209b11df36
Binary files /dev/null and b/trained_models/9191_fournights/run1/models/actor_agent4.pt differ
diff --git a/trained_models/9191_fournights/run1/models/actor_agent5.pt b/trained_models/9191_fournights/run1/models/actor_agent5.pt
new file mode 100644
index 0000000000000000000000000000000000000000..d7ac11d20a30f414e6ac2dfec763604487e83644
Binary files /dev/null and b/trained_models/9191_fournights/run1/models/actor_agent5.pt differ
diff --git a/trained_models/9191_fournights/run1/models/actor_agent6.pt b/trained_models/9191_fournights/run1/models/actor_agent6.pt
new file mode 100644
index 0000000000000000000000000000000000000000..1ef58eb1f7265473ddd9ac3e9a865955b56fcd59
Binary files /dev/null and b/trained_models/9191_fournights/run1/models/actor_agent6.pt differ
diff --git a/trained_models/9191_fournights/run1/models/critic_agent0.pt b/trained_models/9191_fournights/run1/models/critic_agent0.pt
new file mode 100644
index 0000000000000000000000000000000000000000..86f1aae0a32456c59087457d363f9a4f3cba97bd
Binary files /dev/null and b/trained_models/9191_fournights/run1/models/critic_agent0.pt differ
diff --git a/trained_models/9191_fournights/run1/models/critic_agent1.pt b/trained_models/9191_fournights/run1/models/critic_agent1.pt
new file mode 100644
index 0000000000000000000000000000000000000000..882aa316e8b141e5865516c0fceac00233602c03
Binary files /dev/null and b/trained_models/9191_fournights/run1/models/critic_agent1.pt differ
diff --git a/trained_models/9191_fournights/run1/models/critic_agent2.pt b/trained_models/9191_fournights/run1/models/critic_agent2.pt
new file mode 100644
index 0000000000000000000000000000000000000000..66380ca1e225c715ef207159b4d7ee35d9f0fc0b
Binary files /dev/null and b/trained_models/9191_fournights/run1/models/critic_agent2.pt differ
diff --git a/trained_models/9191_fournights/run1/models/critic_agent3.pt b/trained_models/9191_fournights/run1/models/critic_agent3.pt
new file mode 100644
index 0000000000000000000000000000000000000000..39e2e7b6639122a17f123ed34325a62922dae2e2
Binary files /dev/null and b/trained_models/9191_fournights/run1/models/critic_agent3.pt differ
diff --git a/trained_models/9191_fournights/run1/models/critic_agent4.pt b/trained_models/9191_fournights/run1/models/critic_agent4.pt
new file mode 100644
index 0000000000000000000000000000000000000000..feba6db3048c53b90d1ad0ac66387774f87bbec7
Binary files /dev/null and b/trained_models/9191_fournights/run1/models/critic_agent4.pt differ
diff --git a/trained_models/9191_fournights/run1/models/critic_agent5.pt b/trained_models/9191_fournights/run1/models/critic_agent5.pt
new file mode 100644
index 0000000000000000000000000000000000000000..bd0c23c2553b6d5d0ccd036369a5e57bc3d7d33a
Binary files /dev/null and b/trained_models/9191_fournights/run1/models/critic_agent5.pt differ
diff --git a/trained_models/9191_fournights/run1/models/critic_agent6.pt b/trained_models/9191_fournights/run1/models/critic_agent6.pt
new file mode 100644
index 0000000000000000000000000000000000000000..fb7551a659ffa7870636a371ce8276ef054e2454
Binary files /dev/null and b/trained_models/9191_fournights/run1/models/critic_agent6.pt differ