benson V5 har cooket

MLAgents/config/imitation/Crawler.yaml

@@ -0,0 +1,39 @@
+behaviors:
+  Crawler:
+    trainer_type: ppo
+    hyperparameters:
+      batch_size: 2024
+      buffer_size: 20240
+      learning_rate: 0.0003
+      beta: 0.005
+      epsilon: 0.2
+      lambd: 0.95
+      num_epoch: 3
+      learning_rate_schedule: linear
+    network_settings:
+      normalize: true
+      hidden_units: 512
+      num_layers: 3
+      vis_encode_type: simple
+    reward_signals:
+      gail:
+        gamma: 0.99
+        strength: 1.0
+        network_settings:
+          normalize: true
+          hidden_units: 128
+          num_layers: 2
+          vis_encode_type: simple
+        learning_rate: 0.0003
+        use_actions: false
+        use_vail: false
+        demo_path: Project/Assets/ML-Agents/Examples/Crawler/Demos/ExpertCrawler.demo
+    keep_checkpoints: 5
+    max_steps: 10000000
+    time_horizon: 1000
+    summary_freq: 30000
+    behavioral_cloning:
+      demo_path: Project/Assets/ML-Agents/Examples/Crawler/Demos/ExpertCrawler.demo
+      steps: 50000
+      strength: 0.5
+      samples_per_update: 0

MLAgents/config/imitation/Hallway.yaml

@@ -0,0 +1,35 @@
+behaviors:
+  Hallway:
+    trainer_type: ppo
+    hyperparameters:
+      batch_size: 128
+      buffer_size: 1024
+      learning_rate: 0.0003
+      beta: 0.01
+      epsilon: 0.2
+      lambd: 0.95
+      num_epoch: 3
+      learning_rate_schedule: linear
+    network_settings:
+      normalize: false
+      hidden_units: 128
+      num_layers: 2
+      vis_encode_type: simple
+      memory:
+        sequence_length: 64
+        memory_size: 256
+    reward_signals:
+      extrinsic:
+        gamma: 0.99
+        strength: 1.0
+      gail:
+        gamma: 0.99
+        strength: 0.01
+        learning_rate: 0.0003
+        use_actions: false
+        use_vail: false
+        demo_path: Project/Assets/ML-Agents/Examples/Hallway/Demos/ExpertHallway.demo
+    keep_checkpoints: 5
+    max_steps: 10000000
+    time_horizon: 64
+    summary_freq: 10000

MLAgents/config/imitation/PushBlock.yaml

@@ -0,0 +1,42 @@
+behaviors:
+  PushBlock:
+    trainer_type: ppo
+    hyperparameters:
+      batch_size: 128
+      buffer_size: 2048
+      learning_rate: 0.0003
+      beta: 0.01
+      epsilon: 0.2
+      lambd: 0.95
+      num_epoch: 3
+      learning_rate_schedule: linear
+    network_settings:
+      normalize: false
+      hidden_units: 256
+      num_layers: 2
+      vis_encode_type: simple
+    reward_signals:
+      extrinsic:
+        gamma: 0.99
+        strength: 1.0
+      gail:
+        gamma: 0.99
+        strength: 0.01
+        network_settings:
+          normalize: false
+          hidden_units: 128
+          num_layers: 2
+          vis_encode_type: simple
+        learning_rate: 0.0003
+        use_actions: false
+        use_vail: false
+        demo_path: Project/Assets/ML-Agents/Examples/PushBlock/Demos/ExpertPushBlock.demo
+    keep_checkpoints: 5
+    max_steps: 100000
+    time_horizon: 64
+    summary_freq: 60000
+    behavioral_cloning:
+      demo_path: Project/Assets/ML-Agents/Examples/PushBlock/Demos/ExpertPushBlock.demo
+      steps: 50000
+      strength: 1.0
+      samples_per_update: 0

MLAgents/config/imitation/Pyramids.yaml

@@ -0,0 +1,34 @@
+behaviors:
+  Pyramids:
+    trainer_type: ppo
+    time_horizon: 128
+    max_steps: 1.0e7
+    hyperparameters:
+      batch_size: 128
+      beta: 0.01
+      buffer_size: 2048
+      epsilon: 0.2
+      lambd: 0.95
+      learning_rate: 0.0003
+      num_epoch: 3
+    network_settings:
+      num_layers: 2
+      normalize: false
+      hidden_units: 512
+    reward_signals:
+      extrinsic:
+        strength: 1.0
+        gamma: 0.99
+      curiosity:
+        strength: 0.02
+        gamma: 0.99
+        network_settings:
+          hidden_units: 256
+      gail:
+        strength: 0.01
+        gamma: 0.99
+        demo_path: Project/Assets/ML-Agents/Examples/Pyramids/Demos/ExpertPyramid.demo
+    behavioral_cloning:
+      demo_path: Project/Assets/ML-Agents/Examples/Pyramids/Demos/ExpertPyramid.demo
+      strength: 0.5
+      steps: 150000