Matteobettini's workspace
Runs
81
Name
3 visualized
task_name: balance
task_name: balance
9
27
algorithm_name: ippo
algorithm_name: ippo
3
algorithm_name: iddpg
algorithm_name: iddpg
3
algorithm_name: isac
algorithm_name: isac
3
algorithm_name: qmix
algorithm_name: qmix
3
algorithm_name: vdn
algorithm_name: vdn
3
algorithm_name: iql
algorithm_name: iql
3
algorithm_name: maddpg
algorithm_name: maddpg
3
algorithm_name: mappo
algorithm_name: mappo
3
algorithm_name: masac
algorithm_name: masac
3
State
Notes
User
Tags
Created
Runtime
Sweep
algorithm_config.alpha_init
algorithm_config.clip_epsilon
algorithm_config.critic_coef
algorithm_config.delay_qvalue
algorithm_config.delay_value
algorithm_config.discrete_target_entropy_weight
algorithm_config.entropy_coef
algorithm_config.fixed_alpha
algorithm_config.lmbda
algorithm_config.loss_critic_type
algorithm_config.loss_function
algorithm_config.mixing_embed_dim
algorithm_config.num_qvalue_nets
algorithm_config.share_param_critic
algorithm_config.target_entropy
algorithm_name
continuous_actions
environment_name
experiment_config.adam_eps
experiment_config.checkpoint_interval
experiment_config.clip_grad_norm
experiment_config.clip_grad_val
experiment_config.create_json
experiment_config.evaluation
experiment_config.evaluation_episodes
experiment_config.evaluation_interval
experiment_config.exploration_anneal_frames
experiment_config.exploration_eps_end
experiment_config.exploration_eps_init
experiment_config.gamma
experiment_config.hard_target_update_frequency
experiment_config.loggers
experiment_config.lr
experiment_config.max_n_frames
experiment_config.off_policy_collected_frames_per_batch
experiment_config.off_policy_memory_size
experiment_config.off_policy_n_envs_per_worker
experiment_config.off_policy_n_optimizer_steps
experiment_config.off_policy_train_batch_size
experiment_config.on_policy_collected_frames_per_batch
experiment_config.on_policy_minibatch_size
experiment_config.on_policy_n_envs_per_worker
experiment_config.on_policy_n_minibatch_iters
experiment_config.polyak_tau
Finished
-
matteobettini
2d 10h 55m 42s
-
1
0.2
1
true
true
0.2
0
false
0.9
l2
l2
32
2
true
auto
["iddpg","ippo","iql","isac","maddpg","mappo","masac","qmix","vdn"]
[false,true]
vmas
0.000001
300000
true
5
true
true
200
120000
1000000
0.01
0.8
0.9
5
wandb
0.00005
10000000
6000
1000000
60
1000
128
60000
4096
600
45
0.005
Finished
-
matteobettini
1d 4h 46m 5s
-
-
0.2
1
-
-
-
0
-
0.9
l2
-
-
-
true
-
ippo
true
vmas
0.000001
300000
true
5
true
true
200
120000
1000000
0.01
0.8
0.9
5
wandb
0.00005
10000000
6000
1000000
60
1000
128
60000
4096
600
45
0.005
Finished
-
matteobettini
1d 10h 10m 51s
-
-
-
-
-
true
-
-
-
-
-
l2
-
-
true
-
iddpg
true
vmas
0.000001
300000
true
5
true
true
200
120000
1000000
0.01
0.8
0.9
5
wandb
0.00005
10000000
6000
1000000
60
1000
128
60000
4096
600
45
0.005
Finished
-
matteobettini
1d 14h 24s
-
1
-
-
true
-
0.2
-
false
-
-
l2
-
2
true
auto
isac
true
vmas
0.000001
300000
true
5
true
true
200
120000
1000000
0.01
0.8
0.9
5
wandb
0.00005
10000000
6000
1000000
60
1000
128
60000
4096
600
45
0.005
Finished
-
matteobettini
2d 6h 37m 58s
-
-
-
-
-
true
-
-
-
-
-
l2
32
-
-
-
qmix
false
vmas
0.000001
300000
true
5
true
true
200
120000
1000000
0.01
0.8
0.9
5
wandb
0.00005
10000000
6000
1000000
60
1000
128
60000
4096
600
45
0.005
Finished
-
matteobettini
2d 7h 4m 1s
-
-
-
-
-
true
-
-
-
-
-
l2
-
-
-
-
vdn
false
vmas
0.000001
300000
true
5
true
true
200
120000
1000000
0.01
0.8
0.9
5
wandb
0.00005
10000000
6000
1000000
60
1000
128
60000
4096
600
45
0.005
Finished
-
matteobettini
2d 5h 31m 59s
-
-
-
-
-
true
-
-
-
-
-
l2
-
-
-
-
iql
false
vmas
0.000001
300000
true
5
true
true
200
120000
1000000
0.01
0.8
0.9
5
wandb
0.00005
10000000
6000
1000000
60
1000
128
60000
4096
600
45
0.005
Finished
-
matteobettini
2d 7h 22m 10s
-
-
-
-
-
true
-
-
-
-
-
l2
-
-
true
-
maddpg
true
vmas
0.000001
300000
true
5
true
true
200
120000
1000000
0.01
0.8
0.9
5
wandb
0.00005
10000000
6000
1000000
60
1000
128
60000
4096
600
45
0.005
Finished
-
matteobettini
2d 1h 21m 45s
-
-
0.2
1
-
-
-
0
-
0.9
l2
-
-
-
true
-
mappo
true
vmas
0.000001
300000
true
5
true
true
200
120000
1000000
0.01
0.8
0.9
5
wandb
0.00005
10000000
6000
1000000
60
1000
128
60000
4096
600
45
0.005
Finished
-
matteobettini
2d 8h 57m 17s
-
1
-
-
true
-
0.2
-
false
-
-
l2
-
2
true
auto
masac
true
vmas
0.000001
300000
true
5
true
true
200
120000
1000000
0.01
0.8
0.9
5
wandb
0.00005
10000000
6000
1000000
60
1000
128
60000
4096
600
45
0.005
1-1
of 1