Second, improved training of PyramidsRDS with PPO

Pyramids.onnx

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3395dfc05a3eb245da447c483971f7847b23dc78d8ba60e6d7e23333f993321b
+oid sha256:315caf35221e6c229abfb8a3cfc48a204b94f96d74a66fc38239e889edaf8b48
 size 1420238

Pyramids/Pyramids-1000030.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:315caf35221e6c229abfb8a3cfc48a204b94f96d74a66fc38239e889edaf8b48
+size 1420238

Pyramids/Pyramids-1000030.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e1e4cf84038d7aa3c8c4ff3c4d787619b5ad1ee21fcb7f5bc69ee9d5f627c9ec
+size 8656252

Pyramids/Pyramids-499955.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a99372cfce9169c815d0dcea84c1271ac84cabbc467ad50c2372c7ebda3440d4
+size 1420238

Pyramids/Pyramids-499955.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0e1b1b490260e53933a76a213ce6bff1bbafdeb8d60ef616d32f812a6cd48a0a
+size 8656173

Pyramids/Pyramids-999902.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:315caf35221e6c229abfb8a3cfc48a204b94f96d74a66fc38239e889edaf8b48
+size 1420238

Pyramids/Pyramids-999902.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5e03c0ce1ef2f5038b555895acff7c989120c2ad0bd7b7edfaa51694e1eb89af
+size 8656173

Pyramids/checkpoint.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:278a2e9d11e0f557d47c982fad4a0cc22ea05974cf9d1a1c3cd0d4f6152ac321
+oid sha256:cb1e4ad310148fa1e31d0824d72b72f388888adb49ab4e5942dcfa7069c0c029
 size 8655778

Pyramids/events.out.tfevents.1704275893.a76fab4c855c.18344.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f7c5baf706decfc6ea17208680fac0cb2716dd9c69354e37cd125e994f03252e
+size 198670

config.json

@@ -1 +1 @@
-{"default_settings": null, "behaviors": {"Pyramids": {"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, "shared_critic": false, "learning_rate_schedule": "linear", "beta_schedule": "linear", "epsilon_schedule": "linear"}, "checkpoint_interval": 500000, "network_settings": {"normalize": false, "hidden_units": 512, "num_layers": 2, "vis_encode_type": "simple", "memory": null, "goal_conditioning_type": "hyper", "deterministic": false}, "reward_signals": {"extrinsic": {"gamma": 0.95, "strength": 1.0, "network_settings": {"normalize": false, "hidden_units": 128, "num_layers": 2, "vis_encode_type": "simple", "memory": null, "goal_conditioning_type": "hyper", "deterministic": false}}, "rnd": {"gamma": 0.94, "strength": 0.01, "network_settings": {"normalize": false, "hidden_units": 64, "num_layers": 3, "vis_encode_type": "simple", "memory": null, "goal_conditioning_type": "hyper", "deterministic": false}, "learning_rate": 0.0001, "encoding_size": null}}, "init_path": null, "keep_checkpoints": 5, "even_checkpoints": false, "max_steps": 1000000, "time_horizon": 128, "summary_freq": 50000, "threaded": false, "self_play": null, "behavioral_cloning": null}}, "env_settings": {"env_path": "./training-envs-executables/linux/Pyramids/Pyramids", "env_args": null, "base_port": 5005, "num_envs": 1, "num_areas": 1, "timeout_wait": 60, "seed": -1, "max_lifetime_restarts": 10, "restarts_rate_limit_n": 1, "restarts_rate_limit_period_s": 60}, "engine_settings": {"width": 84, "height": 84, "quality_level": 5, "time_scale": 20, "target_frame_rate": -1, "capture_frame_rate": 60, "no_graphics": true, "no_graphics_monitor": false}, "environment_parameters": null, "checkpoint_settings": {"run_id": "Pyramids Training-2", "initialize_from": null, "load_model": false, "resume": false, "force": false, "train_model": false, "inference": false, "results_dir": "results"}, "torch_settings": {"device": null}, "debug": false}
+{"default_settings": null, "behaviors": {"Pyramids": {"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, "shared_critic": false, "learning_rate_schedule": "linear", "beta_schedule": "linear", "epsilon_schedule": "linear"}, "checkpoint_interval": 500000, "network_settings": {"normalize": false, "hidden_units": 512, "num_layers": 2, "vis_encode_type": "simple", "memory": null, "goal_conditioning_type": "hyper", "deterministic": false}, "reward_signals": {"extrinsic": {"gamma": 0.99, "strength": 1.0, "network_settings": {"normalize": false, "hidden_units": 128, "num_layers": 2, "vis_encode_type": "simple", "memory": null, "goal_conditioning_type": "hyper", "deterministic": false}}, "rnd": {"gamma": 0.99, "strength": 0.01, "network_settings": {"normalize": false, "hidden_units": 64, "num_layers": 3, "vis_encode_type": "simple", "memory": null, "goal_conditioning_type": "hyper", "deterministic": false}, "learning_rate": 0.0001, "encoding_size": null}}, "init_path": null, "keep_checkpoints": 5, "even_checkpoints": false, "max_steps": 1000000, "time_horizon": 128, "summary_freq": 50000, "threaded": false, "self_play": null, "behavioral_cloning": null}}, "env_settings": {"env_path": "./training-envs-executables/linux/Pyramids/Pyramids", "env_args": null, "base_port": 5005, "num_envs": 1, "num_areas": 1, "timeout_wait": 60, "seed": -1, "max_lifetime_restarts": 10, "restarts_rate_limit_n": 1, "restarts_rate_limit_period_s": 60}, "engine_settings": {"width": 84, "height": 84, "quality_level": 5, "time_scale": 20, "target_frame_rate": -1, "capture_frame_rate": 60, "no_graphics": true, "no_graphics_monitor": false}, "environment_parameters": null, "checkpoint_settings": {"run_id": "Pyramids Training-3", "initialize_from": null, "load_model": false, "resume": false, "force": false, "train_model": false, "inference": false, "results_dir": "results"}, "torch_settings": {"device": null}, "debug": false}

configuration.yaml

@@ -25,7 +25,7 @@ behaviors:
       deterministic: false
     reward_signals:
       extrinsic:
-        gamma: 0.95
+        gamma: 0.99
         strength: 1.0
         network_settings:
           normalize: false
@@ -36,7 +36,7 @@ behaviors:
           goal_conditioning_type: hyper
           deterministic: false
       rnd:
-        gamma: 0.94
+        gamma: 0.99
         strength: 0.01
         network_settings:
           normalize: false
@@ -79,7 +79,7 @@ engine_settings:
   no_graphics_monitor: false
 environment_parameters: null
 checkpoint_settings:
-  run_id: Pyramids Training-2
+  run_id: Pyramids Training-3
   initialize_from: null
   load_model: false
   resume: false

run_logs/Player-0.log

@@ -1,7 +1,7 @@
 Mono path[0] = '/content/ml-agents/training-envs-executables/linux/Pyramids/Pyramids_Data/Managed'
 Mono config path = '/content/ml-agents/training-envs-executables/linux/Pyramids/Pyramids_Data/MonoBleedingEdge/etc'
 Found 1 interfaces on host : 0) 172.28.0.12
-Multi-casting "[IP] 172.28.0.12 [Port] 55334 [Flags] 2 [Guid] 1589326979 [EditorId] 764847374 [Version] 1048832 [Id] LinuxPlayer(13,172.28.0.12) [Debug] 0 [PackageName] LinuxPlayer [ProjectName] UnityEnvironment" to [225.0.0.222:54997]...
+Multi-casting "[IP] 172.28.0.12 [Port] 55435 [Flags] 2 [Guid] 2233454501 [EditorId] 764847374 [Version] 1048832 [Id] LinuxPlayer(13,172.28.0.12) [Debug] 0 [PackageName] LinuxPlayer [ProjectName] UnityEnvironment" to [225.0.0.222:54997]...
 Preloaded 'lib_burst_generated.so'
 Preloaded 'libgrpc_csharp_ext.x64.so'
 Initialize engine version: 2021.3.5f1 (40eb3a945986)
@@ -33,7 +33,7 @@ ALSA lib pcm.c:2664:(snd_pcm_open_noupdate) Unknown PCM default
 FMOD failed to initialize the output device.: "Error initializing output device. " (60)
 FMOD initialized on nosound output
 Begin MonoManager ReloadAssembly
-- Completed reload, in  0.192 seconds
+- Completed reload, in  0.107 seconds
 ERROR: Shader Sprites/Default shader is not supported on this GPU (none of subshaders/fallbacks are suitable)
 ERROR: Shader Sprites/Mask shader is not supported on this GPU (none of subshaders/fallbacks are suitable)
 ERROR: Shader Legacy Shaders/VertexLit shader is not supported on this GPU (none of subshaders/fallbacks are suitable)
@@ -65,7 +65,7 @@ ERROR: Shader Standard shader is not supported on this GPU (none of subshaders/f
 WARNING: Shader Unsupported: 'Standard' - All subshaders removed
 WARNING: Shader Did you use #pragma only_renderers and omit this platform?
 WARNING: Shader If subshaders removal was intentional, you may have forgotten turning Fallback off?
-UnloadTime: 1.130649 ms
+UnloadTime: 0.874010 ms
 ERROR: Shader UI/Default shader is not supported on this GPU (none of subshaders/fallbacks are suitable)
 requesting resize 84 x 84
 Setting up 1 worker threads for Enlighten.
@@ -73,7 +73,7 @@ PlayerConnection::CleanupMemory Statistics:
 [ALLOC_TEMP_TLS] TLS Allocator
   StackAllocators : 
     [ALLOC_TEMP_MAIN]
-      Peak usage frame count: [16.0 KB-32.0 KB]: 7292 frames, [32.0 KB-64.0 KB]: 10155 frames, [64.0 KB-128.0 KB]: 1317 frames, [2.0 MB-4.0 MB]: 1 frames
+      Peak usage frame count: [16.0 KB-32.0 KB]: 4076 frames, [32.0 KB-64.0 KB]: 12111 frames, [64.0 KB-128.0 KB]: 2578 frames, [2.0 MB-4.0 MB]: 1 frames
       Initial Block Size 4.0 MB
       Current Block Size 4.0 MB
       Peak Allocated Bytes 2.0 MB
@@ -123,17 +123,17 @@ PlayerConnection::CleanupMemory Statistics:
       Current Block Size 64.0 KB
       Peak Allocated Bytes 0 B
       Overflow Count 0
-    [ALLOC_TEMP_Background Job.Worker 15]
+    [ALLOC_TEMP_Background Job.Worker 2]
       Initial Block Size 32.0 KB
       Current Block Size 32.0 KB
       Peak Allocated Bytes 0 B
       Overflow Count 0
-    [ALLOC_TEMP_Background Job.Worker 2]
+    [ALLOC_TEMP_Background Job.Worker 7]
       Initial Block Size 32.0 KB
       Current Block Size 32.0 KB
       Peak Allocated Bytes 0 B
       Overflow Count 0
-    [ALLOC_TEMP_Background Job.Worker 7]
+    [ALLOC_TEMP_Background Job.Worker 15]
       Initial Block Size 32.0 KB
       Current Block Size 32.0 KB
       Peak Allocated Bytes 0 B
@@ -146,7 +146,7 @@ PlayerConnection::CleanupMemory Statistics:
     [ALLOC_TEMP_Profiler.Dispatcher]
       Initial Block Size 64.0 KB
       Current Block Size 64.0 KB
-      Peak Allocated Bytes 240 B
+      Peak Allocated Bytes 480 B
       Overflow Count 0
     [ALLOC_TEMP_Background Job.Worker 12]
       Initial Block Size 32.0 KB
@@ -204,28 +204,28 @@ PlayerConnection::CleanupMemory Statistics:
       Peak Allocated Bytes 128 B
       Overflow Count 0
 [ALLOC_MEMORYPROFILER]
-  Peak usage frame count: [0.5 MB-1.0 MB]: 299 frames, [1.0 MB-2.0 MB]: 18466 frames
+  Peak usage frame count: [0.5 MB-1.0 MB]: 299 frames, [1.0 MB-2.0 MB]: 18467 frames
   Requested Block Size 1.0 MB
   Peak Block count 2
   Peak Allocated memory 1.6 MB
   Peak Large allocation bytes 0 B
 [ALLOC_DEFAULT] Dual Thread Allocator
-  Peak main deferred allocation count 11658
+  Peak main deferred allocation count 13249
     [ALLOC_BUCKET]
       Large Block size 4.0 MB
       Used Block count 1
       Peak Allocated bytes 1.7 MB
     [ALLOC_DEFAULT_MAIN]
-      Peak usage frame count: [8.0 MB-16.0 MB]: 1 frames, [16.0 MB-32.0 MB]: 18764 frames
+      Peak usage frame count: [8.0 MB-16.0 MB]: 1 frames, [16.0 MB-32.0 MB]: 18765 frames
       Requested Block Size 16.0 MB
       Peak Block count 3
-      Peak Allocated memory 30.9 MB
+      Peak Allocated memory 31.3 MB
       Peak Large allocation bytes 0 B
     [ALLOC_DEFAULT_THREAD]
-      Peak usage frame count: [16.0 MB-32.0 MB]: 18765 frames
+      Peak usage frame count: [16.0 MB-32.0 MB]: 18766 frames
       Requested Block Size 16.0 MB
       Peak Block count 1
-      Peak Allocated memory 23.6 MB
+      Peak Allocated memory 22.3 MB
       Peak Large allocation bytes 16.0 MB
 [ALLOC_TEMP_JOB_1_FRAME]
   Initial Block Size 2.0 MB
@@ -254,13 +254,13 @@ PlayerConnection::CleanupMemory Statistics:
       Used Block count 1
       Peak Allocated bytes 1.7 MB
     [ALLOC_GFX_MAIN]
-      Peak usage frame count: [32.0 KB-64.0 KB]: 18764 frames, [64.0 KB-128.0 KB]: 1 frames
+      Peak usage frame count: [32.0 KB-64.0 KB]: 18765 frames, [64.0 KB-128.0 KB]: 1 frames
       Requested Block Size 16.0 MB
       Peak Block count 1
       Peak Allocated memory 67.0 KB
       Peak Large allocation bytes 0 B
     [ALLOC_GFX_THREAD]
-      Peak usage frame count: [64.0 KB-128.0 KB]: 18765 frames
+      Peak usage frame count: [64.0 KB-128.0 KB]: 18766 frames
       Requested Block Size 16.0 MB
       Peak Block count 1
       Peak Allocated memory 67.7 KB
@@ -272,16 +272,16 @@ PlayerConnection::CleanupMemory Statistics:
       Used Block count 1
       Peak Allocated bytes 1.7 MB
     [ALLOC_CACHEOBJECTS_MAIN]
-      Peak usage frame count: [0.5 MB-1.0 MB]: 299 frames, [1.0 MB-2.0 MB]: 18466 frames
+      Peak usage frame count: [0.5 MB-1.0 MB]: 299 frames, [1.0 MB-2.0 MB]: 18467 frames
       Requested Block Size 4.0 MB
       Peak Block count 1
       Peak Allocated memory 1.4 MB
       Peak Large allocation bytes 0 B
     [ALLOC_CACHEOBJECTS_THREAD]
-      Peak usage frame count: [1.0 MB-2.0 MB]: 18764 frames, [4.0 MB-8.0 MB]: 1 frames
+      Peak usage frame count: [1.0 MB-2.0 MB]: 18765 frames, [4.0 MB-8.0 MB]: 1 frames
       Requested Block Size 4.0 MB
       Peak Block count 2
-      Peak Allocated memory 4.6 MB
+      Peak Allocated memory 4.7 MB
       Peak Large allocation bytes 0 B
 [ALLOC_TYPETREE] Dual Thread Allocator
   Peak main deferred allocation count 0
@@ -290,25 +290,25 @@ PlayerConnection::CleanupMemory Statistics:
       Used Block count 1
       Peak Allocated bytes 1.7 MB
     [ALLOC_TYPETREE_MAIN]
-      Peak usage frame count: [0-1.0 KB]: 18765 frames
+      Peak usage frame count: [0-1.0 KB]: 18766 frames
       Requested Block Size 2.0 MB
       Peak Block count 1
       Peak Allocated memory 1.0 KB
       Peak Large allocation bytes 0 B
     [ALLOC_TYPETREE_THREAD]
-      Peak usage frame count: [1.0 KB-2.0 KB]: 18765 frames
+      Peak usage frame count: [1.0 KB-2.0 KB]: 18766 frames
       Requested Block Size 2.0 MB
       Peak Block count 1
       Peak Allocated memory 1.9 KB
       Peak Large allocation bytes 0 B
 [ALLOC_PROFILER]
-  Peak usage frame count: [16.0 KB-32.0 KB]: 18765 frames
+  Peak usage frame count: [16.0 KB-32.0 KB]: 18766 frames
   Requested Block Size 16.0 MB
   Peak Block count 1
-  Peak Allocated memory 29.2 KB
+  Peak Allocated memory 29.1 KB
   Peak Large allocation bytes 0 B
     [ALLOC_PROFILER_BUCKET]
       Large Block size 4.0 MB
       Used Block count 1
       Peak Allocated bytes 396 B
-##utp:{"type":"MemoryLeaks","version":2,"phase":"Immediate","time":1704275352635,"processId":7898,"allocatedMemory":1970133,"memoryLabels":[{"Default":9033},{"Permanent":1264},{"NewDelete":12801},{"Thread":34460},{"Manager":10603},{"VertexData":12},{"Geometry":280},{"Texture":16},{"Shader":69173},{"Material":24},{"GfxDevice":35248},{"Animation":304},{"Audio":3976},{"Physics":288},{"Serialization":216},{"Input":9176},{"JobScheduler":200},{"Mono":40},{"ScriptingNativeRuntime":216},{"BaseObject":1609212},{"Resource":592},{"Renderer":1936},{"Transform":48},{"File":800},{"WebCam":24},{"Culling":40},{"Terrain":953},{"Wind":24},{"String":3447},{"DynamicArray":30868},{"HashMap":7680},{"Utility":1360},{"PoolAlloc":1160},{"TypeTree":1792},{"ScriptManager":80},{"RuntimeInitializeOnLoadManager":72},{"SpriteAtlas":112},{"GI":3272},{"Unet":16},{"Director":7760},{"WebRequest":720},{"VR":45473},{"SceneManager":424},{"Video":32},{"LazyScriptCache":32},{"NativeArray":384},{"Camera":25},{"Secure":1},{"SerializationCache":624},{"APIUpdating":5872},{"Subsystems":384},{"VirtualTexturing":57552},{"AssetReference":32}]}
+##utp:{"type":"MemoryLeaks","version":2,"phase":"Immediate","time":1704278030062,"processId":18361,"allocatedMemory":1970133,"memoryLabels":[{"Default":9033},{"Permanent":1264},{"NewDelete":12801},{"Thread":34460},{"Manager":10603},{"VertexData":12},{"Geometry":280},{"Texture":16},{"Shader":69173},{"Material":24},{"GfxDevice":35248},{"Animation":304},{"Audio":3976},{"Physics":288},{"Serialization":216},{"Input":9176},{"JobScheduler":200},{"Mono":40},{"ScriptingNativeRuntime":216},{"BaseObject":1609212},{"Resource":592},{"Renderer":1936},{"Transform":48},{"File":800},{"WebCam":24},{"Culling":40},{"Terrain":953},{"Wind":24},{"String":3447},{"DynamicArray":30868},{"HashMap":7680},{"Utility":1360},{"PoolAlloc":1160},{"TypeTree":1792},{"ScriptManager":80},{"RuntimeInitializeOnLoadManager":72},{"SpriteAtlas":112},{"GI":3272},{"Unet":16},{"Director":7760},{"WebRequest":720},{"VR":45473},{"SceneManager":424},{"Video":32},{"LazyScriptCache":32},{"NativeArray":384},{"Camera":25},{"Secure":1},{"SerializationCache":624},{"APIUpdating":5872},{"Subsystems":384},{"VirtualTexturing":57552},{"AssetReference":32}]}

run_logs/timers.json

@@ -2,171 +2,171 @@
     "name": "root",
     "gauges": {
         "Pyramids.Policy.Entropy.mean": {
-            "value": 0.5983927845954895,
-            "min": 0.5983927845954895,
-            "max": 1.343400001525879,
+            "value": 0.40660443902015686,
+            "min": 0.3838774263858795,
+            "max": 1.3838720321655273,
             "count": 20
         },
         "Pyramids.Policy.Entropy.sum": {
-            "value": 29986.66015625,
-            "min": 29986.66015625,
-            "max": 67234.484375,
+            "value": 20323.716796875,
+            "min": 19058.74609375,
+            "max": 70632.828125,
             "count": 20
         },
         "Pyramids.Step.mean": {
-            "value": 999925.0,
-            "min": 49878.0,
-            "max": 999925.0,
+            "value": 999902.0,
+            "min": 49982.0,
+            "max": 999902.0,
             "count": 20
         },
         "Pyramids.Step.sum": {
-            "value": 999925.0,
-            "min": 49878.0,
-            "max": 999925.0,
+            "value": 999902.0,
+            "min": 49982.0,
+            "max": 999902.0,
             "count": 20
         },
         "Pyramids.Policy.ExtrinsicValueEstimate.mean": {
-            "value": -0.013876122422516346,
-            "min": -0.030839893966913223,
-            "max": -0.00945535209029913,
+            "value": 0.5106868147850037,
+            "min": -0.10902372002601624,
+            "max": 0.5106868147850037,
             "count": 20
         },
         "Pyramids.Policy.ExtrinsicValueEstimate.sum": {
-            "value": -5.5782012939453125,
-            "min": -12.3051176071167,
-            "max": -3.81050705909729,
+            "value": 230.8304443359375,
+            "min": -43.60948944091797,
+            "max": 230.8304443359375,
             "count": 20
         },
         "Pyramids.Policy.RndValueEstimate.mean": {
-            "value": 0.0029050002340227365,
-            "min": -0.00022533306037075818,
-            "max": 0.08890791237354279,
+            "value": 0.018002863973379135,
+            "min": 0.017526134848594666,
+            "max": 0.27170729637145996,
             "count": 20
         },
         "Pyramids.Policy.RndValueEstimate.sum": {
-            "value": 1.167810082435608,
-            "min": -0.09035855531692505,
-            "max": 35.47425842285156,
+            "value": 8.13729476928711,
+            "min": 7.298983573913574,
+            "max": 108.68292236328125,
             "count": 20
         },
         "Pyramids.Losses.PolicyLoss.mean": {
-            "value": 0.06579226099424805,
-            "min": 0.06512104197606153,
-            "max": 0.07199981855150664,
+            "value": 0.0690960958714177,
+            "min": 0.06685300014880363,
+            "max": 0.07348646662135302,
             "count": 20
         },
         "Pyramids.Losses.PolicyLoss.sum": {
-            "value": 1.513222002867705,
-            "min": 0.8930364629323959,
-            "max": 1.6477607740883151,
+            "value": 1.6583063009140246,
+            "min": 1.0288105326989423,
+            "max": 1.7034627192473386,
             "count": 20
         },
         "Pyramids.Losses.ValueLoss.mean": {
-            "value": 0.0009892530458551413,
-            "min": 0.0001602431029824168,
-            "max": 0.0037391359524955254,
+            "value": 0.015689400035530914,
+            "min": 0.0004058652203409273,
+            "max": 0.017001819814943114,
             "count": 20
         },
         "Pyramids.Losses.ValueLoss.sum": {
-            "value": 0.02275282005466825,
-            "min": 0.0035253482656131697,
-            "max": 0.04860876738244183,
+            "value": 0.3765456008527419,
+            "min": 0.008697785586991082,
+            "max": 0.40804367555863474,
             "count": 20
         },
         "Pyramids.Policy.LearningRate.mean": {
-            "value": 7.53210618498261e-06,
-            "min": 7.53210618498261e-06,
-            "max": 0.0002917810642781077,
+            "value": 7.360097546666664e-06,
+            "min": 7.360097546666664e-06,
+            "max": 0.00029159310280229997,
             "count": 20
         },
         "Pyramids.Policy.LearningRate.sum": {
-            "value": 0.00017323844225460003,
-            "min": 0.00017323844225460003,
-            "max": 0.005768409877196799,
+            "value": 0.00017664234111999994,
+            "min": 0.00017664234111999994,
+            "max": 0.0054513657828781995,
             "count": 20
         },
         "Pyramids.Policy.Epsilon.mean": {
-            "value": 0.1025106695652174,
-            "min": 0.1025106695652174,
-            "max": 0.19726035384615387,
+            "value": 0.10245333333333334,
+            "min": 0.10245333333333334,
+            "max": 0.1971977,
             "count": 20
         },
         "Pyramids.Policy.Epsilon.sum": {
-            "value": 2.3577454,
-            "min": 2.3577454,
-            "max": 4.1228032,
+            "value": 2.45888,
+            "min": 2.45888,
+            "max": 4.0171218,
             "count": 20
         },
         "Pyramids.Policy.Beta.mean": {
-            "value": 0.00026081588956521746,
-            "min": 0.00026081588956521746,
-            "max": 0.00972630934923077,
+            "value": 0.00025508799999999994,
+            "min": 0.00025508799999999994,
+            "max": 0.00972005023,
             "count": 20
         },
         "Pyramids.Policy.Beta.sum": {
-            "value": 0.005998765460000001,
-            "min": 0.005998765460000001,
-            "max": 0.19230803968000001,
+            "value": 0.006122111999999999,
+            "min": 0.006122111999999999,
+            "max": 0.18175046781999998,
             "count": 20
         },
         "Pyramids.Losses.RNDLoss.mean": {
-            "value": 0.01551153976470232,
-            "min": 0.01551153976470232,
-            "max": 0.3107236623764038,
+            "value": 0.015770716592669487,
+            "min": 0.015770716592669487,
+            "max": 0.37413766980171204,
             "count": 20
         },
         "Pyramids.Losses.RNDLoss.sum": {
-            "value": 0.35676541924476624,
-            "min": 0.35676541924476624,
-            "max": 4.039407730102539,
+            "value": 0.3784972131252289,
+            "min": 0.3784972131252289,
+            "max": 5.237927436828613,
             "count": 20
         },
         "Pyramids.Environment.EpisodeLength.mean": {
-            "value": 972.5098039215686,
-            "min": 946.2692307692307,
-            "max": 996.0416666666666,
+            "value": 355.95620437956205,
+            "min": 355.95620437956205,
+            "max": 999.0,
             "count": 20
         },
         "Pyramids.Environment.EpisodeLength.sum": {
-            "value": 49598.0,
-            "min": 47237.0,
-            "max": 56043.0,
+            "value": 48766.0,
+            "min": 47053.0,
+            "max": 52985.0,
             "count": 20
         },
         "Pyramids.Environment.CumulativeReward.mean": {
-            "value": -0.7378627939843664,
-            "min": -0.9552083844318986,
-            "max": -0.6432923601510433,
+            "value": 1.5564291726281172,
+            "min": -1.0000000521540642,
+            "max": 1.5564291726281172,
             "count": 20
         },
         "Pyramids.Environment.CumulativeReward.sum": {
-            "value": -37.631002493202686,
-            "min": -50.088402941823006,
-            "max": -33.45120272785425,
+            "value": 213.23079665005207,
+            "min": -48.00000250339508,
+            "max": 213.23079665005207,
             "count": 20
         },
         "Pyramids.Policy.ExtrinsicReward.mean": {
-            "value": -0.7378627939843664,
-            "min": -0.9552083844318986,
-            "max": -0.6432923601510433,
+            "value": 1.5564291726281172,
+            "min": -1.0000000521540642,
+            "max": 1.5564291726281172,
             "count": 20
         },
         "Pyramids.Policy.ExtrinsicReward.sum": {
-            "value": -37.631002493202686,
-            "min": -50.088402941823006,
-            "max": -33.45120272785425,
+            "value": 213.23079665005207,
+            "min": -48.00000250339508,
+            "max": 213.23079665005207,
             "count": 20
         },
         "Pyramids.Policy.RndReward.mean": {
-            "value": 0.15703592533428296,
-            "min": 0.15624511372656202,
-            "max": 4.615633514864991,
+            "value": 0.057838770586890624,
+            "min": 0.057838770586890624,
+            "max": 5.367142829267929,
             "count": 20
         },
         "Pyramids.Policy.RndReward.sum": {
-            "value": 8.00883219204843,
-            "min": 8.00883219204843,
-            "max": 221.55040871351957,
+            "value": 7.923911570404016,
+            "min": 7.923911570404016,
+            "max": 257.6228558048606,
             "count": 20
         },
         "Pyramids.IsTraining.mean": {
@@ -184,67 +184,67 @@
     },
     "metadata": {
         "timer_format_version": "0.1.0",
-        "start_time_seconds": "1704273359",
+        "start_time_seconds": "1704275892",
         "python_version": "3.10.12 (main, Nov 20 2023, 15:14:05) [GCC 11.4.0]",
-        "command_line_arguments": "/usr/local/bin/mlagents-learn ./config/ppo/PyramidsRND.yaml --env=./training-envs-executables/linux/Pyramids/Pyramids --run-id=Pyramids Training-2 --no-graphics",
+        "command_line_arguments": "/usr/local/bin/mlagents-learn ./config/ppo/PyramidsRND.yaml --env=./training-envs-executables/linux/Pyramids/Pyramids --run-id=Pyramids Training-3 --no-graphics",
         "mlagents_version": "1.1.0.dev0",
         "mlagents_envs_version": "1.1.0.dev0",
         "communication_protocol_version": "1.5.0",
         "pytorch_version": "2.1.2+cu121",
         "numpy_version": "1.23.5",
-        "end_time_seconds": "1704275352"
+        "end_time_seconds": "1704278030"
     },
-    "total": 1993.154926618,
+    "total": 2138.0222330259994,
     "count": 1,
-    "self": 0.48554877099991245,
+    "self": 0.9853462019991639,
     "children": {
         "run_training.setup": {
-            "total": 0.06937176099995668,
+            "total": 0.04736289700031193,
             "count": 1,
-            "self": 0.06937176099995668
+            "self": 0.04736289700031193
         },
         "TrainerController.start_learning": {
-            "total": 1992.6000060860001,
+            "total": 2136.989523927,
             "count": 1,
-            "self": 1.4480502969940972,
+            "self": 1.458595546028846,
             "children": {
                 "TrainerController._reset_env": {
-                    "total": 2.673110766000036,
+                    "total": 2.147141593000015,
                     "count": 1,
-                    "self": 2.673110766000036
+                    "self": 2.147141593000015
                 },
                 "TrainerController.advance": {
-                    "total": 1988.394409740006,
-                    "count": 63030,
-                    "self": 1.4689806919566308,
+                    "total": 2133.255490715972,
+                    "count": 63628,
+                    "self": 1.5977552797594399,
                     "children": {
                         "env_step": {
-                            "total": 1358.9817631640058,
-                            "count": 63030,
-                            "self": 1225.942064266986,
+                            "total": 1490.3827759572332,
+                            "count": 63628,
+                            "self": 1354.185206475171,
                             "children": {
                                 "SubprocessEnvManager._take_step": {
-                                    "total": 132.17192505101366,
-                                    "count": 63030,
-                                    "self": 4.719373328042593,
+                                    "total": 135.26424939209664,
+                                    "count": 63628,
+                                    "self": 4.834128640190102,
                                     "children": {
                                         "TorchPolicy.evaluate": {
-                                            "total": 127.45255172297107,
-                                            "count": 62550,
-                                            "self": 127.45255172297107
+                                            "total": 130.43012075190654,
+                                            "count": 62551,
+                                            "self": 130.43012075190654
                                         }
                                     }
                                 },
                                 "workers": {
-                                    "total": 0.8677738460060027,
-                                    "count": 63030,
+                                    "total": 0.9333200899654912,
+                                    "count": 63628,
                                     "self": 0.0,
                                     "children": {
                                         "worker_root": {
-                                            "total": 1987.4621567540394,
-                                            "count": 63030,
+                                            "total": 2131.672288137017,
+                                            "count": 63628,
                                             "is_parallel": true,
-                                            "self": 880.5503308880463,
+                                            "self": 900.1139222690681,
                                             "children": {
                                                 "run_training.setup": {
                                                     "total": 0.0,
@@ -253,48 +253,48 @@
                                                     "self": 0.0,
                                                     "children": {
                                                         "steps_from_proto": {
-                                                            "total": 0.0022598160001052747,
+                                                            "total": 0.0017603739997866796,
                                                             "count": 1,
                                                             "is_parallel": true,
-                                                            "self": 0.0005843530000220198,
+                                                            "self": 0.0005314810023264727,
                                                             "children": {
                                                                 "_process_rank_one_or_two_observation": {
-                                                                    "total": 0.0016754630000832549,
+                                                                    "total": 0.001228892997460207,
                                                                     "count": 8,
                                                                     "is_parallel": true,
-                                                                    "self": 0.0016754630000832549
+                                                                    "self": 0.001228892997460207
                                                                 }
                                                             }
                                                         },
                                                         "UnityEnvironment.step": {
-                                                            "total": 0.05077610300008928,
+                                                            "total": 0.04862397000033525,
                                                             "count": 1,
                                                             "is_parallel": true,
-                                                            "self": 0.0006098310000197671,
+                                                            "self": 0.0005931800005782861,
                                                             "children": {
                                                                 "UnityEnvironment._generate_step_input": {
-                                                                    "total": 0.00048121800000444637,
+                                                                    "total": 0.0004881089998889365,
                                                                     "count": 1,
                                                                     "is_parallel": true,
-                                                                    "self": 0.00048121800000444637
+                                                                    "self": 0.0004881089998889365
                                                                 },
                                                                 "communicator.exchange": {
-                                                                    "total": 0.047941176999984236,
+                                                                    "total": 0.04590963299961004,
                                                                     "count": 1,
                                                                     "is_parallel": true,
-                                                                    "self": 0.047941176999984236
+                                                                    "self": 0.04590963299961004
                                                                 },
                                                                 "steps_from_proto": {
-                                                                    "total": 0.001743877000080829,
+                                                                    "total": 0.0016330480002579861,
                                                                     "count": 1,
                                                                     "is_parallel": true,
-                                                                    "self": 0.00037161100021876337,
+                                                                    "self": 0.00035718700019060634,
                                                                     "children": {
                                                                         "_process_rank_one_or_two_observation": {
-                                                                            "total": 0.0013722659998620657,
+                                                                            "total": 0.0012758610000673798,
                                                                             "count": 8,
                                                                             "is_parallel": true,
-                                                                            "self": 0.0013722659998620657
+                                                                            "self": 0.0012758610000673798
                                                                         }
                                                                     }
                                                                 }
@@ -303,34 +303,34 @@
                                                     }
                                                 },
                                                 "UnityEnvironment.step": {
-                                                    "total": 1106.911825865993,
-                                                    "count": 63029,
+                                                    "total": 1231.5583658679489,
+                                                    "count": 63627,
                                                     "is_parallel": true,
-                                                    "self": 34.95583143394447,
+                                                    "self": 34.945650922931236,
                                                     "children": {
                                                         "UnityEnvironment._generate_step_input": {
-                                                            "total": 24.578581023000424,
-                                                            "count": 63029,
+                                                            "total": 24.486377486137826,
+                                                            "count": 63627,
                                                             "is_parallel": true,
-                                                            "self": 24.578581023000424
+                                                            "self": 24.486377486137826
                                                         },
                                                         "communicator.exchange": {
-                                                            "total": 948.7428121660562,
-                                                            "count": 63029,
+                                                            "total": 1073.1361114249412,
+                                                            "count": 63627,
                                                             "is_parallel": true,
-                                                            "self": 948.7428121660562
+                                                            "self": 1073.1361114249412
                                                         },
                                                         "steps_from_proto": {
-                                                            "total": 98.63460124299195,
-                                                            "count": 63029,
+                                                            "total": 98.9902260339386,
+                                                            "count": 63627,
                                                             "is_parallel": true,
-                                                            "self": 19.71788833404821,
+                                                            "self": 20.087128791734358,
                                                             "children": {
                                                                 "_process_rank_one_or_two_observation": {
-                                                                    "total": 78.91671290894374,
-                                                                    "count": 504232,
+                                                                    "total": 78.90309724220424,
+                                                                    "count": 509016,
                                                                     "is_parallel": true,
-                                                                    "self": 78.91671290894374
+                                                                    "self": 78.90309724220424
                                                                 }
                                                             }
                                                         }
@@ -343,31 +343,31 @@
                             }
                         },
                         "trainer_advance": {
-                            "total": 627.9436658840436,
-                            "count": 63030,
-                            "self": 2.6710124700555298,
+                            "total": 641.2749594789793,
+                            "count": 63628,
+                            "self": 2.883373502001632,
                             "children": {
                                 "process_trajectory": {
-                                    "total": 124.30694764398822,
-                                    "count": 63030,
-                                    "self": 124.06981375898795,
+                                    "total": 128.47727592296815,
+                                    "count": 63628,
+                                    "self": 128.2043659059682,
                                     "children": {
                                         "RLTrainer._checkpoint": {
-                                            "total": 0.23713388500027577,
+                                            "total": 0.2729100169999583,
                                             "count": 2,
-                                            "self": 0.23713388500027577
+                                            "self": 0.2729100169999583
                                         }
                                     }
                                 },
                                 "_update_policy": {
-                                    "total": 500.9657057699999,
-                                    "count": 442,
-                                    "self": 298.7058735060057,
+                                    "total": 509.91431005400955,
+                                    "count": 447,
+                                    "self": 302.8242494070273,
                                     "children": {
                                         "TorchPPOOptimizer.update": {
-                                            "total": 202.2598322639942,
-                                            "count": 22710,
-                                            "self": 202.2598322639942
+                                            "total": 207.09006064698224,
+                                            "count": 22893,
+                                            "self": 207.09006064698224
                                         }
                                     }
                                 }
@@ -376,19 +376,19 @@
                     }
                 },
                 "trainer_threads": {
-                    "total": 8.469996828353032e-07,
+                    "total": 1.2289992810110562e-06,
                     "count": 1,
-                    "self": 8.469996828353032e-07
+                    "self": 1.2289992810110562e-06
                 },
                 "TrainerController._save_models": {
-                    "total": 0.08443443600026512,
+                    "total": 0.1282948429998214,
                     "count": 1,
-                    "self": 0.001465669000026537,
+                    "self": 0.002003808999688772,
                     "children": {
                         "RLTrainer._checkpoint": {
-                            "total": 0.08296876700023859,
+                            "total": 0.12629103400013264,
                             "count": 1,
-                            "self": 0.08296876700023859
+                            "self": 0.12629103400013264
                         }
                     }
                 }

run_logs/training_status.json

@@ -2,40 +2,40 @@
     "Pyramids": {
         "checkpoints": [
             {
-                "steps": 499925,
-                "file_path": "results/Pyramids Training-2/Pyramids/Pyramids-499925.onnx",
-                "reward": -0.9996000528335571,
-                "creation_time": 1704274353.265112,
+                "steps": 499955,
+                "file_path": "results/Pyramids Training-3/Pyramids/Pyramids-499955.onnx",
+                "reward": null,
+                "creation_time": 1704276891.3628292,
                 "auxillary_file_paths": [
-                    "results/Pyramids Training-2/Pyramids/Pyramids-499925.pt"
+                    "results/Pyramids Training-3/Pyramids/Pyramids-499955.pt"
                 ]
             },
             {
-                "steps": 999925,
-                "file_path": "results/Pyramids Training-2/Pyramids/Pyramids-999925.onnx",
-                "reward": 0.16050000116229057,
-                "creation_time": 1704275352.109694,
+                "steps": 999902,
+                "file_path": "results/Pyramids Training-3/Pyramids/Pyramids-999902.onnx",
+                "reward": null,
+                "creation_time": 1704278029.0157154,
                 "auxillary_file_paths": [
-                    "results/Pyramids Training-2/Pyramids/Pyramids-999925.pt"
+                    "results/Pyramids Training-3/Pyramids/Pyramids-999902.pt"
                 ]
             },
             {
-                "steps": 1000053,
-                "file_path": "results/Pyramids Training-2/Pyramids/Pyramids-1000053.onnx",
-                "reward": 0.16050000116229057,
-                "creation_time": 1704275352.2048194,
+                "steps": 1000030,
+                "file_path": "results/Pyramids Training-3/Pyramids/Pyramids-1000030.onnx",
+                "reward": null,
+                "creation_time": 1704278029.1617928,
                 "auxillary_file_paths": [
-                    "results/Pyramids Training-2/Pyramids/Pyramids-1000053.pt"
+                    "results/Pyramids Training-3/Pyramids/Pyramids-1000030.pt"
                 ]
             }
         ],
         "final_checkpoint": {
-            "steps": 1000053,
-            "file_path": "results/Pyramids Training-2/Pyramids.onnx",
-            "reward": 0.16050000116229057,
-            "creation_time": 1704275352.2048194,
+            "steps": 1000030,
+            "file_path": "results/Pyramids Training-3/Pyramids.onnx",
+            "reward": null,
+            "creation_time": 1704278029.1617928,
             "auxillary_file_paths": [
-                "results/Pyramids Training-2/Pyramids/Pyramids-1000053.pt"
+                "results/Pyramids Training-3/Pyramids/Pyramids-1000030.pt"
             ]
         }
     },