benson V5 har cooket

MLAgents/config/sac/3DBall.yaml

@@ -0,0 +1,27 @@
+behaviors:
+  3DBall:
+    trainer_type: sac
+    hyperparameters:
+      learning_rate: 0.0003
+      learning_rate_schedule: constant
+      batch_size: 64
+      buffer_size: 200000
+      buffer_init_steps: 0
+      tau: 0.005
+      steps_per_update: 10.0
+      save_replay_buffer: false
+      init_entcoef: 0.5
+      reward_signal_steps_per_update: 10.0
+    network_settings:
+      normalize: true
+      hidden_units: 64
+      num_layers: 2
+      vis_encode_type: simple
+    reward_signals:
+      extrinsic:
+        gamma: 0.99
+        strength: 1.0
+    keep_checkpoints: 5
+    max_steps: 200000
+    time_horizon: 1000
+    summary_freq: 12000

MLAgents/config/sac/3DBallHard.yaml

@@ -0,0 +1,27 @@
+behaviors:
+  3DBallHard:
+    trainer_type: sac
+    hyperparameters:
+      learning_rate: 0.0003
+      learning_rate_schedule: constant
+      batch_size: 256
+      buffer_size: 500000
+      buffer_init_steps: 0
+      tau: 0.005
+      steps_per_update: 10.0
+      save_replay_buffer: false
+      init_entcoef: 1.0
+      reward_signal_steps_per_update: 10.0
+    network_settings:
+      normalize: true
+      hidden_units: 128
+      num_layers: 2
+      vis_encode_type: simple
+    reward_signals:
+      extrinsic:
+        gamma: 0.99
+        strength: 1.0
+    keep_checkpoints: 5
+    max_steps: 500000
+    time_horizon: 1000
+    summary_freq: 12000

MLAgents/config/sac/Basic.yaml

@@ -0,0 +1,27 @@
+behaviors:
+  Basic:
+    trainer_type: sac
+    hyperparameters:
+      learning_rate: 0.0003
+      learning_rate_schedule: constant
+      batch_size: 64
+      buffer_size: 50000
+      buffer_init_steps: 0
+      tau: 0.005
+      steps_per_update: 10.0
+      save_replay_buffer: false
+      init_entcoef: 0.01
+      reward_signal_steps_per_update: 10.0
+    network_settings:
+      normalize: false
+      hidden_units: 20
+      num_layers: 2
+      vis_encode_type: simple
+    reward_signals:
+      extrinsic:
+        gamma: 0.99
+        strength: 1.0
+    keep_checkpoints: 5
+    max_steps: 500000
+    time_horizon: 10
+    summary_freq: 2000

MLAgents/config/sac/Crawler.yaml

@@ -0,0 +1,27 @@
+behaviors:
+  Crawler:
+    trainer_type: sac
+    hyperparameters:
+      learning_rate: 0.0003
+      learning_rate_schedule: constant
+      batch_size: 256
+      buffer_size: 500000
+      buffer_init_steps: 0
+      tau: 0.005
+      steps_per_update: 20.0
+      save_replay_buffer: false
+      init_entcoef: 1.0
+      reward_signal_steps_per_update: 20.0
+    network_settings:
+      normalize: true
+      hidden_units: 512
+      num_layers: 3
+      vis_encode_type: simple
+    reward_signals:
+      extrinsic:
+        gamma: 0.995
+        strength: 1.0
+    keep_checkpoints: 5
+    max_steps: 5000000
+    time_horizon: 1000
+    summary_freq: 30000

MLAgents/config/sac/FoodCollector.yaml

@@ -0,0 +1,28 @@
+behaviors:
+  FoodCollector:
+    trainer_type: sac
+    hyperparameters:
+      learning_rate: 0.0003
+      learning_rate_schedule: constant
+      batch_size: 256
+      buffer_size: 2048
+      buffer_init_steps: 0
+      tau: 0.005
+      steps_per_update: 10.0
+      save_replay_buffer: false
+      init_entcoef: 0.05
+      reward_signal_steps_per_update: 10.0
+    network_settings:
+      normalize: false
+      hidden_units: 256
+      num_layers: 1
+      vis_encode_type: simple
+    reward_signals:
+      extrinsic:
+        gamma: 0.99
+        strength: 1.0
+    keep_checkpoints: 5
+    max_steps: 2000000
+    time_horizon: 64
+    summary_freq: 60000
+    threaded: false

MLAgents/config/sac/GridWorld.yaml

@@ -0,0 +1,27 @@
+behaviors:
+  GridWorld:
+    trainer_type: sac
+    hyperparameters:
+      learning_rate: 0.0003
+      learning_rate_schedule: constant
+      batch_size: 128
+      buffer_size: 50000
+      buffer_init_steps: 1000
+      tau: 0.005
+      steps_per_update: 10.0
+      save_replay_buffer: false
+      init_entcoef: 0.5
+      reward_signal_steps_per_update: 10.0
+    network_settings:
+      normalize: false
+      hidden_units: 128
+      num_layers: 1
+      vis_encode_type: simple
+    reward_signals:
+      extrinsic:
+        gamma: 0.9
+        strength: 1.0
+    keep_checkpoints: 5
+    max_steps: 500000
+    time_horizon: 5
+    summary_freq: 20000

MLAgents/config/sac/Hallway.yaml

@@ -0,0 +1,30 @@
+behaviors:
+  Hallway:
+    trainer_type: sac
+    hyperparameters:
+      learning_rate: 0.0003
+      learning_rate_schedule: constant
+      batch_size: 512
+      buffer_size: 200000
+      buffer_init_steps: 0
+      tau: 0.005
+      steps_per_update: 10.0
+      save_replay_buffer: false
+      init_entcoef: 0.1
+      reward_signal_steps_per_update: 10.0
+    network_settings:
+      normalize: false
+      hidden_units: 128
+      num_layers: 2
+      vis_encode_type: simple
+      memory:
+        sequence_length: 64
+        memory_size: 128
+    reward_signals:
+      extrinsic:
+        gamma: 0.99
+        strength: 1.0
+    keep_checkpoints: 5
+    max_steps: 4000000
+    time_horizon: 64
+    summary_freq: 10000

MLAgents/config/sac/PushBlock.yaml

@@ -0,0 +1,27 @@
+behaviors:
+  PushBlock:
+    trainer_type: sac
+    hyperparameters:
+      learning_rate: 0.0003
+      learning_rate_schedule: constant
+      batch_size: 128
+      buffer_size: 50000
+      buffer_init_steps: 0
+      tau: 0.005
+      steps_per_update: 10.0
+      save_replay_buffer: false
+      init_entcoef: 0.05
+      reward_signal_steps_per_update: 10.0
+    network_settings:
+      normalize: false
+      hidden_units: 256
+      num_layers: 2
+      vis_encode_type: simple
+    reward_signals:
+      extrinsic:
+        gamma: 0.99
+        strength: 1.0
+    keep_checkpoints: 5
+    max_steps: 2000000
+    time_horizon: 64
+    summary_freq: 100000

MLAgents/config/sac/Pyramids.yaml

@@ -0,0 +1,34 @@
+behaviors:
+  Pyramids:
+    trainer_type: sac
+    hyperparameters:
+      learning_rate: 0.0003
+      learning_rate_schedule: constant
+      batch_size: 128
+      buffer_size: 2000000
+      buffer_init_steps: 1000
+      tau: 0.01
+      steps_per_update: 10.0
+      save_replay_buffer: false
+      init_entcoef: 0.01
+      reward_signal_steps_per_update: 10.0
+    network_settings:
+      normalize: false
+      hidden_units: 512
+      num_layers: 3
+      vis_encode_type: simple
+    reward_signals:
+      extrinsic:
+        gamma: 0.995
+        strength: 2.0
+      gail:
+        gamma: 0.99
+        strength: 0.01
+        learning_rate: 0.0003
+        use_actions: true
+        use_vail: false
+        demo_path: Project/Assets/ML-Agents/Examples/Pyramids/Demos/ExpertPyramid.demo
+    keep_checkpoints: 5
+    max_steps: 3000000
+    time_horizon: 128
+    summary_freq: 30000

MLAgents/config/sac/Walker.yaml

@@ -0,0 +1,27 @@
+behaviors:
+  Walker:
+    trainer_type: sac
+    hyperparameters:
+      learning_rate: 0.0003
+      learning_rate_schedule: constant
+      batch_size: 1024
+      buffer_size: 2000000
+      buffer_init_steps: 0
+      tau: 0.005
+      steps_per_update: 30.0
+      save_replay_buffer: false
+      init_entcoef: 1.0
+      reward_signal_steps_per_update: 30.0
+    network_settings:
+      normalize: true
+      hidden_units: 256
+      num_layers: 3
+      vis_encode_type: simple
+    reward_signals:
+      extrinsic:
+        gamma: 0.995
+        strength: 1.0
+    keep_checkpoints: 5
+    max_steps: 15000000
+    time_horizon: 1000
+    summary_freq: 30000

MLAgents/config/sac/WallJump.yaml

@@ -0,0 +1,53 @@
+behaviors:
+  BigWallJump:
+    trainer_type: sac
+    hyperparameters:
+      learning_rate: 0.0003
+      learning_rate_schedule: constant
+      batch_size: 128
+      buffer_size: 200000
+      buffer_init_steps: 0
+      tau: 0.005
+      steps_per_update: 20.0
+      save_replay_buffer: false
+      init_entcoef: 0.1
+      reward_signal_steps_per_update: 10.0
+    network_settings:
+      normalize: false
+      hidden_units: 256
+      num_layers: 2
+      vis_encode_type: simple
+    reward_signals:
+      extrinsic:
+        gamma: 0.99
+        strength: 1.0
+    keep_checkpoints: 5
+    max_steps: 15000000
+    time_horizon: 128
+    summary_freq: 20000
+  SmallWallJump:
+    trainer_type: sac
+    hyperparameters:
+      learning_rate: 0.0003
+      learning_rate_schedule: constant
+      batch_size: 128
+      buffer_size: 50000
+      buffer_init_steps: 0
+      tau: 0.005
+      steps_per_update: 20.0
+      save_replay_buffer: false
+      init_entcoef: 0.1
+      reward_signal_steps_per_update: 10.0
+    network_settings:
+      normalize: false
+      hidden_units: 256
+      num_layers: 2
+      vis_encode_type: simple
+    reward_signals:
+      extrinsic:
+        gamma: 0.99
+        strength: 1.0
+    keep_checkpoints: 5
+    max_steps: 5000000
+    time_horizon: 128
+    summary_freq: 20000

MLAgents/config/sac/Worm.yaml

@@ -0,0 +1,27 @@
+behaviors:
+  Worm:
+    trainer_type: sac
+    hyperparameters:
+      learning_rate: 0.0003
+      learning_rate_schedule: constant
+      batch_size: 256
+      buffer_size: 500000
+      buffer_init_steps: 0
+      tau: 0.005
+      steps_per_update: 20.0
+      save_replay_buffer: false
+      init_entcoef: 1.0
+      reward_signal_steps_per_update: 20.0
+    network_settings:
+      normalize: true
+      hidden_units: 512
+      num_layers: 3
+      vis_encode_type: simple
+    reward_signals:
+      extrinsic:
+        gamma: 0.995
+        strength: 1.0
+    keep_checkpoints: 5
+    max_steps: 5000000
+    time_horizon: 1000
+    summary_freq: 30000