VAZaytsev commited on
Commit
099632f
1 Parent(s): f890465
README.md ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+
2
+ ---
3
+ tags:
4
+ - unity-ml-agents
5
+ - ml-agents
6
+ - deep-reinforcement-learning
7
+ - reinforcement-learning
8
+ - ML-Agents-SnowballTarget
9
+ library_name: ml-agents
10
+ ---
11
+
12
+ # **ppo** Agent playing **SnowballTarget**
13
+ This is a trained model of a **ppo** agent playing **SnowballTarget** using the [Unity ML-Agents Library](https://github.com/Unity-Technologies/ml-agents).
14
+
15
+ ## Usage (with ML-Agents)
16
+ The Documentation: https://github.com/huggingface/ml-agents#get-started
17
+ We wrote a complete tutorial to learn to train your first agent using ML-Agents and publish it to the Hub:
18
+
19
+
20
+ ### Resume the training
21
+ ```
22
+ mlagents-learn <your_configuration_file_path.yaml> --run-id=<run_id> --resume
23
+ ```
24
+ ### Watch your Agent play
25
+ You can watch your agent **playing directly in your browser:**.
26
+
27
+ 1. Go to https://huggingface.co/spaces/unity/ML-Agents-SnowballTarget
28
+ 2. Step 1: Write your model_id: VAZaytsev/Reinforce-SnowballTarget1
29
+ 3. Step 2: Select your *.nn /*.onnx file
30
+ 4. Click on Watch the agent play 👀
31
+
SnowballTarget.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7ebfcc0633fc6ec9b42f0486fd9db22e537d97c513b79e8f7d70756157b94ed1
3
+ size 645119
SnowballTarget/SnowballTarget-149984.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4e8ac06167db79cf2abb59da9b3505cdd9667d7db04cb589dd3a9cfb35dd3153
3
+ size 645119
SnowballTarget/SnowballTarget-149984.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d6991aab66af2f5722ec9b5e5aab324e783d8b783a2b4accd90da4ba90c7c914
3
+ size 3845584
SnowballTarget/SnowballTarget-199984.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7ebfcc0633fc6ec9b42f0486fd9db22e537d97c513b79e8f7d70756157b94ed1
3
+ size 645119
SnowballTarget/SnowballTarget-199984.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:313e874b2fdd4427619522d1e9a1774ee7845b27be9ee0e9c98c39b2ecfe80cf
3
+ size 3845584
SnowballTarget/SnowballTarget-200112.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7ebfcc0633fc6ec9b42f0486fd9db22e537d97c513b79e8f7d70756157b94ed1
3
+ size 645119
SnowballTarget/SnowballTarget-200112.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:43ff57787a7ebe88e9b0631add242e804868688858107c369e3c4a4fd2f78d6b
3
+ size 3845584
SnowballTarget/SnowballTarget-49936.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5663f0aa87e2f24ce95bff058750f785b5dc63182eb3e1e9d626f38500d88d04
3
+ size 645119
SnowballTarget/SnowballTarget-49936.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:da126acea90bb776dbff3ea2927e6c315b8cd4e692a466f2cdc1adb6ef4ce6da
3
+ size 3845584
SnowballTarget/SnowballTarget-99960.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:215e49d6fe6e5e47e21613b000866ee9f9fa9bcbe593682e2f645a448cfaa450
3
+ size 645119
SnowballTarget/SnowballTarget-99960.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f4a87847557531075524b89621f9f37a96f52ae1d9789ee579fcd1b7fe84f952
3
+ size 3845584
SnowballTarget/checkpoint.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:43ff57787a7ebe88e9b0631add242e804868688858107c369e3c4a4fd2f78d6b
3
+ size 3845584
SnowballTarget/events.out.tfevents.1678019826.a8f1827c770e.1524.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4f83f56d4f97d2b6b4a9eb93914d68deb86c5ce4800562519a307a616f2348de
3
+ size 23749
config.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"default_settings": null, "behaviors": {"SnowballTarget": {"trainer_type": "ppo", "hyperparameters": {"batch_size": 128, "buffer_size": 2048, "learning_rate": 0.0003, "beta": 0.005, "epsilon": 0.2, "lambd": 0.95, "num_epoch": 3, "learning_rate_schedule": "linear", "beta_schedule": "linear", "epsilon_schedule": "linear"}, "network_settings": {"normalize": false, "hidden_units": 256, "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}}}, "init_path": null, "keep_checkpoints": 10, "checkpoint_interval": 50000, "max_steps": 200000, "time_horizon": 64, "summary_freq": 10000, "threaded": true, "self_play": null, "behavioral_cloning": null}}, "env_settings": {"env_path": "./training-envs-executables/linux/SnowballTarget/SnowballTarget", "env_args": null, "base_port": 5005, "num_envs": 1, "num_areas": 1, "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}, "environment_parameters": null, "checkpoint_settings": {"run_id": "SnowballTarget1", "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 ADDED
@@ -0,0 +1,75 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ default_settings: null
2
+ behaviors:
3
+ SnowballTarget:
4
+ trainer_type: ppo
5
+ hyperparameters:
6
+ batch_size: 128
7
+ buffer_size: 2048
8
+ learning_rate: 0.0003
9
+ beta: 0.005
10
+ epsilon: 0.2
11
+ lambd: 0.95
12
+ num_epoch: 3
13
+ learning_rate_schedule: linear
14
+ beta_schedule: linear
15
+ epsilon_schedule: linear
16
+ network_settings:
17
+ normalize: false
18
+ hidden_units: 256
19
+ num_layers: 2
20
+ vis_encode_type: simple
21
+ memory: null
22
+ goal_conditioning_type: hyper
23
+ deterministic: false
24
+ reward_signals:
25
+ extrinsic:
26
+ gamma: 0.99
27
+ strength: 1.0
28
+ network_settings:
29
+ normalize: false
30
+ hidden_units: 128
31
+ num_layers: 2
32
+ vis_encode_type: simple
33
+ memory: null
34
+ goal_conditioning_type: hyper
35
+ deterministic: false
36
+ init_path: null
37
+ keep_checkpoints: 10
38
+ checkpoint_interval: 50000
39
+ max_steps: 200000
40
+ time_horizon: 64
41
+ summary_freq: 10000
42
+ threaded: true
43
+ self_play: null
44
+ behavioral_cloning: null
45
+ env_settings:
46
+ env_path: ./training-envs-executables/linux/SnowballTarget/SnowballTarget
47
+ env_args: null
48
+ base_port: 5005
49
+ num_envs: 1
50
+ num_areas: 1
51
+ seed: -1
52
+ max_lifetime_restarts: 10
53
+ restarts_rate_limit_n: 1
54
+ restarts_rate_limit_period_s: 60
55
+ engine_settings:
56
+ width: 84
57
+ height: 84
58
+ quality_level: 5
59
+ time_scale: 20
60
+ target_frame_rate: -1
61
+ capture_frame_rate: 60
62
+ no_graphics: true
63
+ environment_parameters: null
64
+ checkpoint_settings:
65
+ run_id: SnowballTarget1
66
+ initialize_from: null
67
+ load_model: false
68
+ resume: false
69
+ force: false
70
+ train_model: false
71
+ inference: false
72
+ results_dir: results
73
+ torch_settings:
74
+ device: null
75
+ debug: false
run_logs/Player-0.log ADDED
@@ -0,0 +1,262 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ Mono path[0] = '/content/ml-agents/training-envs-executables/linux/SnowballTarget/SnowballTarget_Data/Managed'
2
+ Mono config path = '/content/ml-agents/training-envs-executables/linux/SnowballTarget/SnowballTarget_Data/MonoBleedingEdge/etc'
3
+ Preloaded 'lib_burst_generated.so'
4
+ Preloaded 'libgrpc_csharp_ext.x64.so'
5
+ PlayerPrefs - Creating folder: /root/.config/unity3d/Hugging Face
6
+ PlayerPrefs - Creating folder: /root/.config/unity3d/Hugging Face/SnowballTarget
7
+ Unable to load player prefs
8
+ Initialize engine version: 2021.3.14f1 (eee1884e7226)
9
+ [Subsystems] Discovering subsystems at path /content/ml-agents/training-envs-executables/linux/SnowballTarget/SnowballTarget_Data/UnitySubsystems
10
+ Forcing GfxDevice: Null
11
+ GfxDevice: creating device client; threaded=0; jobified=0
12
+ NullGfxDevice:
13
+ Version: NULL 1.0 [1.0]
14
+ Renderer: Null Device
15
+ Vendor: Unity Technologies
16
+ ALSA lib confmisc.c:767:(parse_card) cannot find card '0'
17
+ ALSA lib conf.c:4732:(_snd_config_evaluate) function snd_func_card_driver returned error: No such file or directory
18
+ ALSA lib confmisc.c:392:(snd_func_concat) error evaluating strings
19
+ ALSA lib conf.c:4732:(_snd_config_evaluate) function snd_func_concat returned error: No such file or directory
20
+ ALSA lib confmisc.c:1246:(snd_func_refer) error evaluating name
21
+ ALSA lib conf.c:4732:(_snd_config_evaluate) function snd_func_refer returned error: No such file or directory
22
+ ALSA lib conf.c:5220:(snd_config_expand) Evaluate error: No such file or directory
23
+ ALSA lib pcm.c:2642:(snd_pcm_open_noupdate) Unknown PCM default
24
+ FMOD failed to initialize the output device.: "Error initializing output device. " (60)
25
+ Forced to initialize FMOD to to the device driver's system output rate 48000, this may impact performance and/or give inconsistent experiences compared to selected sample rate 48000
26
+ ALSA lib confmisc.c:767:(parse_card) cannot find card '0'
27
+ ALSA lib conf.c:4732:(_snd_config_evaluate) function snd_func_card_driver returned error: No such file or directory
28
+ ALSA lib confmisc.c:392:(snd_func_concat) error evaluating strings
29
+ ALSA lib conf.c:4732:(_snd_config_evaluate) function snd_func_concat returned error: No such file or directory
30
+ ALSA lib confmisc.c:1246:(snd_func_refer) error evaluating name
31
+ ALSA lib conf.c:4732:(_snd_config_evaluate) function snd_func_refer returned error: No such file or directory
32
+ ALSA lib conf.c:5220:(snd_config_expand) Evaluate error: No such file or directory
33
+ ALSA lib pcm.c:2642:(snd_pcm_open_noupdate) Unknown PCM default
34
+ FMOD failed to initialize the output device.: "Error initializing output device. " (60)
35
+ FMOD initialized on nosound output
36
+ Begin MonoManager ReloadAssembly
37
+ - Completed reload, in 0.091 seconds
38
+ ERROR: Shader Sprites/Default shader is not supported on this GPU (none of subshaders/fallbacks are suitable)
39
+ ERROR: Shader Sprites/Mask shader is not supported on this GPU (none of subshaders/fallbacks are suitable)
40
+ ERROR: Shader Legacy Shaders/VertexLit shader is not supported on this GPU (none of subshaders/fallbacks are suitable)
41
+ WARNING: Shader Unsupported: 'Standard' - All subshaders removed
42
+ WARNING: Shader Did you use #pragma only_renderers and omit this platform?
43
+ WARNING: Shader If subshaders removal was intentional, you may have forgotten turning Fallback off?
44
+ ERROR: Shader Standard shader is not supported on this GPU (none of subshaders/fallbacks are suitable)
45
+ WARNING: Shader Unsupported: 'Standard' - All subshaders removed
46
+ WARNING: Shader Did you use #pragma only_renderers and omit this platform?
47
+ WARNING: Shader If subshaders removal was intentional, you may have forgotten turning Fallback off?
48
+ UnloadTime: 0.705442 ms
49
+ ERROR: Shader UI/Default shader is not supported on this GPU (none of subshaders/fallbacks are suitable)
50
+ requesting resize 84 x 84
51
+ Setting up 1 worker threads for Enlighten.
52
+ Memory Statistics:
53
+ [ALLOC_TEMP_TLS] TLS Allocator
54
+ StackAllocators :
55
+ [ALLOC_TEMP_MAIN]
56
+ Peak usage frame count: [8.0 KB-16.0 KB]: 5374 frames, [16.0 KB-32.0 KB]: 91 frames, [2.0 MB-4.0 MB]: 1 frames
57
+ Initial Block Size 4.0 MB
58
+ Current Block Size 4.0 MB
59
+ Peak Allocated Bytes 2.0 MB
60
+ Overflow Count 0
61
+ [ALLOC_TEMP_Loading.AsyncRead]
62
+ Initial Block Size 64.0 KB
63
+ Current Block Size 64.0 KB
64
+ Peak Allocated Bytes 198 B
65
+ Overflow Count 0
66
+ [ALLOC_TEMP_Loading.PreloadManager]
67
+ Initial Block Size 256.0 KB
68
+ Current Block Size 304.0 KB
69
+ Peak Allocated Bytes 228.9 KB
70
+ Overflow Count 4
71
+ [ALLOC_TEMP_Background Job.Worker 8]
72
+ Initial Block Size 32.0 KB
73
+ Current Block Size 32.0 KB
74
+ Peak Allocated Bytes 0 B
75
+ Overflow Count 0
76
+ [ALLOC_TEMP_Background Job.Worker 9]
77
+ Initial Block Size 32.0 KB
78
+ Current Block Size 32.0 KB
79
+ Peak Allocated Bytes 0 B
80
+ Overflow Count 0
81
+ [ALLOC_TEMP_Job.Worker 0]
82
+ Initial Block Size 256.0 KB
83
+ Current Block Size 256.0 KB
84
+ Peak Allocated Bytes 3.6 KB
85
+ Overflow Count 0
86
+ [ALLOC_TEMP_Background Job.Worker 10]
87
+ Initial Block Size 32.0 KB
88
+ Current Block Size 32.0 KB
89
+ Peak Allocated Bytes 0 B
90
+ Overflow Count 0
91
+ [ALLOC_TEMP_Background Job.Worker 14]
92
+ Initial Block Size 32.0 KB
93
+ Current Block Size 32.0 KB
94
+ Peak Allocated Bytes 0 B
95
+ Overflow Count 0
96
+ [ALLOC_TEMP_Background Job.Worker 6]
97
+ Initial Block Size 32.0 KB
98
+ Current Block Size 32.0 KB
99
+ Peak Allocated Bytes 0 B
100
+ Overflow Count 0
101
+ [ALLOC_TEMP_Background Job.Worker 12]
102
+ Initial Block Size 32.0 KB
103
+ Current Block Size 32.0 KB
104
+ Peak Allocated Bytes 0 B
105
+ Overflow Count 0
106
+ [ALLOC_TEMP_EnlightenWorker]
107
+ Initial Block Size 64.0 KB
108
+ Current Block Size 64.0 KB
109
+ Peak Allocated Bytes 0 B
110
+ Overflow Count 0
111
+ [ALLOC_TEMP_Background Job.Worker 15]
112
+ Initial Block Size 32.0 KB
113
+ Current Block Size 32.0 KB
114
+ Peak Allocated Bytes 0 B
115
+ Overflow Count 0
116
+ [ALLOC_TEMP_Background Job.Worker 1]
117
+ Initial Block Size 32.0 KB
118
+ Current Block Size 32.0 KB
119
+ Peak Allocated Bytes 0 B
120
+ Overflow Count 0
121
+ [ALLOC_TEMP_Background Job.Worker 2]
122
+ Initial Block Size 32.0 KB
123
+ Current Block Size 32.0 KB
124
+ Peak Allocated Bytes 0 B
125
+ Overflow Count 0
126
+ [ALLOC_TEMP_Background Job.Worker 7]
127
+ Initial Block Size 32.0 KB
128
+ Current Block Size 32.0 KB
129
+ Peak Allocated Bytes 0 B
130
+ Overflow Count 0
131
+ [ALLOC_TEMP_AssetGarbageCollectorHelper]
132
+ Initial Block Size 64.0 KB
133
+ Current Block Size 64.0 KB
134
+ Peak Allocated Bytes 0 B
135
+ Overflow Count 0
136
+ [ALLOC_TEMP_Background Job.Worker 5]
137
+ Initial Block Size 32.0 KB
138
+ Current Block Size 32.0 KB
139
+ Peak Allocated Bytes 0 B
140
+ Overflow Count 0
141
+ [ALLOC_TEMP_Background Job.Worker 13]
142
+ Initial Block Size 32.0 KB
143
+ Current Block Size 32.0 KB
144
+ Peak Allocated Bytes 0 B
145
+ Overflow Count 0
146
+ [ALLOC_TEMP_Background Job.Worker 11]
147
+ Initial Block Size 32.0 KB
148
+ Current Block Size 32.0 KB
149
+ Peak Allocated Bytes 0 B
150
+ Overflow Count 0
151
+ [ALLOC_TEMP_Background Job.Worker 3]
152
+ Initial Block Size 32.0 KB
153
+ Current Block Size 32.0 KB
154
+ Peak Allocated Bytes 0 B
155
+ Overflow Count 0
156
+ [ALLOC_TEMP_Background Job.Worker 0]
157
+ Initial Block Size 32.0 KB
158
+ Current Block Size 32.0 KB
159
+ Peak Allocated Bytes 0 B
160
+ Overflow Count 0
161
+ [ALLOC_TEMP_Background Job.Worker 4]
162
+ Initial Block Size 32.0 KB
163
+ Current Block Size 32.0 KB
164
+ Peak Allocated Bytes 0 B
165
+ Overflow Count 0
166
+ [ALLOC_TEMP_BatchDeleteObjects]
167
+ Initial Block Size 64.0 KB
168
+ Current Block Size 64.0 KB
169
+ Peak Allocated Bytes 0 B
170
+ Overflow Count 0
171
+ [ALLOC_DEFAULT] Dual Thread Allocator
172
+ Peak main deferred allocation count 40
173
+ [ALLOC_BUCKET]
174
+ Large Block size 4.0 MB
175
+ Used Block count 1
176
+ Peak Allocated bytes 1.0 MB
177
+ [ALLOC_DEFAULT_MAIN]
178
+ Peak usage frame count: [4.0 MB-8.0 MB]: 5466 frames
179
+ Requested Block Size 16.0 MB
180
+ Peak Block count 1
181
+ Peak Allocated memory 6.7 MB
182
+ Peak Large allocation bytes 0 B
183
+ [ALLOC_DEFAULT_THREAD]
184
+ Peak usage frame count: [16.0 MB-32.0 MB]: 5466 frames
185
+ Requested Block Size 16.0 MB
186
+ Peak Block count 1
187
+ Peak Allocated memory 17.8 MB
188
+ Peak Large allocation bytes 16.0 MB
189
+ [ALLOC_TEMP_JOB_1_FRAME]
190
+ Initial Block Size 2.0 MB
191
+ Used Block Count 1
192
+ Overflow Count (too large) 0
193
+ Overflow Count (full) 0
194
+ [ALLOC_TEMP_JOB_2_FRAMES]
195
+ Initial Block Size 2.0 MB
196
+ Used Block Count 1
197
+ Overflow Count (too large) 0
198
+ Overflow Count (full) 0
199
+ [ALLOC_TEMP_JOB_4_FRAMES (JobTemp)]
200
+ Initial Block Size 2.0 MB
201
+ Used Block Count 2
202
+ Overflow Count (too large) 0
203
+ Overflow Count (full) 0
204
+ [ALLOC_TEMP_JOB_ASYNC (Background)]
205
+ Initial Block Size 1.0 MB
206
+ Used Block Count 3
207
+ Overflow Count (too large) 0
208
+ Overflow Count (full) 0
209
+ [ALLOC_GFX] Dual Thread Allocator
210
+ Peak main deferred allocation count 0
211
+ [ALLOC_BUCKET]
212
+ Large Block size 4.0 MB
213
+ Used Block count 1
214
+ Peak Allocated bytes 1.0 MB
215
+ [ALLOC_GFX_MAIN]
216
+ Peak usage frame count: [32.0 KB-64.0 KB]: 4337 frames, [64.0 KB-128.0 KB]: 1129 frames
217
+ Requested Block Size 16.0 MB
218
+ Peak Block count 1
219
+ Peak Allocated memory 67.7 KB
220
+ Peak Large allocation bytes 0 B
221
+ [ALLOC_GFX_THREAD]
222
+ Peak usage frame count: [32.0 KB-64.0 KB]: 5466 frames
223
+ Requested Block Size 16.0 MB
224
+ Peak Block count 1
225
+ Peak Allocated memory 39.6 KB
226
+ Peak Large allocation bytes 0 B
227
+ [ALLOC_CACHEOBJECTS] Dual Thread Allocator
228
+ Peak main deferred allocation count 0
229
+ [ALLOC_BUCKET]
230
+ Large Block size 4.0 MB
231
+ Used Block count 1
232
+ Peak Allocated bytes 1.0 MB
233
+ [ALLOC_CACHEOBJECTS_MAIN]
234
+ Peak usage frame count: [0.5 MB-1.0 MB]: 5466 frames
235
+ Requested Block Size 4.0 MB
236
+ Peak Block count 1
237
+ Peak Allocated memory 0.6 MB
238
+ Peak Large allocation bytes 0 B
239
+ [ALLOC_CACHEOBJECTS_THREAD]
240
+ Peak usage frame count: [0.5 MB-1.0 MB]: 5465 frames, [2.0 MB-4.0 MB]: 1 frames
241
+ Requested Block Size 4.0 MB
242
+ Peak Block count 1
243
+ Peak Allocated memory 2.2 MB
244
+ Peak Large allocation bytes 0 B
245
+ [ALLOC_TYPETREE] Dual Thread Allocator
246
+ Peak main deferred allocation count 0
247
+ [ALLOC_BUCKET]
248
+ Large Block size 4.0 MB
249
+ Used Block count 1
250
+ Peak Allocated bytes 1.0 MB
251
+ [ALLOC_TYPETREE_MAIN]
252
+ Peak usage frame count: [0-1.0 KB]: 5466 frames
253
+ Requested Block Size 2.0 MB
254
+ Peak Block count 1
255
+ Peak Allocated memory 1.0 KB
256
+ Peak Large allocation bytes 0 B
257
+ [ALLOC_TYPETREE_THREAD]
258
+ Peak usage frame count: [1.0 KB-2.0 KB]: 5466 frames
259
+ Requested Block Size 2.0 MB
260
+ Peak Block count 1
261
+ Peak Allocated memory 1.7 KB
262
+ Peak Large allocation bytes 0 B
run_logs/timers.json ADDED
@@ -0,0 +1,384 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "name": "root",
3
+ "gauges": {
4
+ "SnowballTarget.Policy.Entropy.mean": {
5
+ "value": 1.051272988319397,
6
+ "min": 1.0170701742172241,
7
+ "max": 2.868896961212158,
8
+ "count": 20
9
+ },
10
+ "SnowballTarget.Policy.Entropy.sum": {
11
+ "value": 10118.501953125,
12
+ "min": 10118.501953125,
13
+ "max": 29380.373046875,
14
+ "count": 20
15
+ },
16
+ "SnowballTarget.Step.mean": {
17
+ "value": 199984.0,
18
+ "min": 9952.0,
19
+ "max": 199984.0,
20
+ "count": 20
21
+ },
22
+ "SnowballTarget.Step.sum": {
23
+ "value": 199984.0,
24
+ "min": 9952.0,
25
+ "max": 199984.0,
26
+ "count": 20
27
+ },
28
+ "SnowballTarget.Policy.ExtrinsicValueEstimate.mean": {
29
+ "value": 12.81018352508545,
30
+ "min": 0.405639111995697,
31
+ "max": 12.81018352508545,
32
+ "count": 20
33
+ },
34
+ "SnowballTarget.Policy.ExtrinsicValueEstimate.sum": {
35
+ "value": 2497.98583984375,
36
+ "min": 78.69398498535156,
37
+ "max": 2590.296875,
38
+ "count": 20
39
+ },
40
+ "SnowballTarget.Environment.EpisodeLength.mean": {
41
+ "value": 199.0,
42
+ "min": 199.0,
43
+ "max": 199.0,
44
+ "count": 20
45
+ },
46
+ "SnowballTarget.Environment.EpisodeLength.sum": {
47
+ "value": 10945.0,
48
+ "min": 8756.0,
49
+ "max": 10945.0,
50
+ "count": 20
51
+ },
52
+ "SnowballTarget.Losses.PolicyLoss.mean": {
53
+ "value": 0.0776061617185655,
54
+ "min": 0.05710873881174598,
55
+ "max": 0.0776061617185655,
56
+ "count": 20
57
+ },
58
+ "SnowballTarget.Losses.PolicyLoss.sum": {
59
+ "value": 0.310424646874262,
60
+ "min": 0.22843495524698393,
61
+ "max": 0.36538541470922525,
62
+ "count": 20
63
+ },
64
+ "SnowballTarget.Losses.ValueLoss.mean": {
65
+ "value": 0.20772941483586443,
66
+ "min": 0.1339574681718231,
67
+ "max": 0.2973157909570956,
68
+ "count": 20
69
+ },
70
+ "SnowballTarget.Losses.ValueLoss.sum": {
71
+ "value": 0.8309176593434577,
72
+ "min": 0.5358298726872924,
73
+ "max": 1.3194364104785172,
74
+ "count": 20
75
+ },
76
+ "SnowballTarget.Policy.LearningRate.mean": {
77
+ "value": 8.082097306000005e-06,
78
+ "min": 8.082097306000005e-06,
79
+ "max": 0.000291882002706,
80
+ "count": 20
81
+ },
82
+ "SnowballTarget.Policy.LearningRate.sum": {
83
+ "value": 3.232838922400002e-05,
84
+ "min": 3.232838922400002e-05,
85
+ "max": 0.00138516003828,
86
+ "count": 20
87
+ },
88
+ "SnowballTarget.Policy.Epsilon.mean": {
89
+ "value": 0.10269400000000001,
90
+ "min": 0.10269400000000001,
91
+ "max": 0.19729400000000002,
92
+ "count": 20
93
+ },
94
+ "SnowballTarget.Policy.Epsilon.sum": {
95
+ "value": 0.41077600000000003,
96
+ "min": 0.41077600000000003,
97
+ "max": 0.96172,
98
+ "count": 20
99
+ },
100
+ "SnowballTarget.Policy.Beta.mean": {
101
+ "value": 0.0001444306000000001,
102
+ "min": 0.0001444306000000001,
103
+ "max": 0.0048649706,
104
+ "count": 20
105
+ },
106
+ "SnowballTarget.Policy.Beta.sum": {
107
+ "value": 0.0005777224000000004,
108
+ "min": 0.0005777224000000004,
109
+ "max": 0.023089828,
110
+ "count": 20
111
+ },
112
+ "SnowballTarget.Environment.CumulativeReward.mean": {
113
+ "value": 25.568181818181817,
114
+ "min": 3.5454545454545454,
115
+ "max": 25.568181818181817,
116
+ "count": 20
117
+ },
118
+ "SnowballTarget.Environment.CumulativeReward.sum": {
119
+ "value": 1125.0,
120
+ "min": 156.0,
121
+ "max": 1396.0,
122
+ "count": 20
123
+ },
124
+ "SnowballTarget.Policy.ExtrinsicReward.mean": {
125
+ "value": 25.568181818181817,
126
+ "min": 3.5454545454545454,
127
+ "max": 25.568181818181817,
128
+ "count": 20
129
+ },
130
+ "SnowballTarget.Policy.ExtrinsicReward.sum": {
131
+ "value": 1125.0,
132
+ "min": 156.0,
133
+ "max": 1396.0,
134
+ "count": 20
135
+ },
136
+ "SnowballTarget.IsTraining.mean": {
137
+ "value": 1.0,
138
+ "min": 1.0,
139
+ "max": 1.0,
140
+ "count": 20
141
+ },
142
+ "SnowballTarget.IsTraining.sum": {
143
+ "value": 1.0,
144
+ "min": 1.0,
145
+ "max": 1.0,
146
+ "count": 20
147
+ }
148
+ },
149
+ "metadata": {
150
+ "timer_format_version": "0.1.0",
151
+ "start_time_seconds": "1678019819",
152
+ "python_version": "3.8.10 (default, Nov 14 2022, 12:59:47) \n[GCC 9.4.0]",
153
+ "command_line_arguments": "/usr/local/bin/mlagents-learn ./config/ppo/SnowballTarget.yaml --env=./training-envs-executables/linux/SnowballTarget/SnowballTarget --run-id=SnowballTarget1 --no-graphics",
154
+ "mlagents_version": "0.29.0.dev0",
155
+ "mlagents_envs_version": "0.29.0.dev0",
156
+ "communication_protocol_version": "1.5.0",
157
+ "pytorch_version": "1.8.1+cu102",
158
+ "numpy_version": "1.22.4",
159
+ "end_time_seconds": "1678020298"
160
+ },
161
+ "total": 478.66905639,
162
+ "count": 1,
163
+ "self": 0.7381102689999466,
164
+ "children": {
165
+ "run_training.setup": {
166
+ "total": 0.11909876900000427,
167
+ "count": 1,
168
+ "self": 0.11909876900000427
169
+ },
170
+ "TrainerController.start_learning": {
171
+ "total": 477.81184735200003,
172
+ "count": 1,
173
+ "self": 0.5789546669935817,
174
+ "children": {
175
+ "TrainerController._reset_env": {
176
+ "total": 9.88233825499998,
177
+ "count": 1,
178
+ "self": 9.88233825499998
179
+ },
180
+ "TrainerController.advance": {
181
+ "total": 467.1611741310064,
182
+ "count": 18219,
183
+ "self": 0.2826362190085092,
184
+ "children": {
185
+ "env_step": {
186
+ "total": 466.8785379119979,
187
+ "count": 18219,
188
+ "self": 324.2312912619899,
189
+ "children": {
190
+ "SubprocessEnvManager._take_step": {
191
+ "total": 142.35295613701027,
192
+ "count": 18219,
193
+ "self": 1.5714029710130148,
194
+ "children": {
195
+ "TorchPolicy.evaluate": {
196
+ "total": 140.78155316599725,
197
+ "count": 18219,
198
+ "self": 31.852579622995222,
199
+ "children": {
200
+ "TorchPolicy.sample_actions": {
201
+ "total": 108.92897354300203,
202
+ "count": 18219,
203
+ "self": 108.92897354300203
204
+ }
205
+ }
206
+ }
207
+ }
208
+ },
209
+ "workers": {
210
+ "total": 0.29429051299774756,
211
+ "count": 18219,
212
+ "self": 0.0,
213
+ "children": {
214
+ "worker_root": {
215
+ "total": 476.14166978599167,
216
+ "count": 18219,
217
+ "is_parallel": true,
218
+ "self": 229.88090772499817,
219
+ "children": {
220
+ "run_training.setup": {
221
+ "total": 0.0,
222
+ "count": 0,
223
+ "is_parallel": true,
224
+ "self": 0.0,
225
+ "children": {
226
+ "steps_from_proto": {
227
+ "total": 0.006295904000012342,
228
+ "count": 1,
229
+ "is_parallel": true,
230
+ "self": 0.005002111000010245,
231
+ "children": {
232
+ "_process_rank_one_or_two_observation": {
233
+ "total": 0.001293793000002097,
234
+ "count": 10,
235
+ "is_parallel": true,
236
+ "self": 0.001293793000002097
237
+ }
238
+ }
239
+ },
240
+ "UnityEnvironment.step": {
241
+ "total": 0.0394459789999928,
242
+ "count": 1,
243
+ "is_parallel": true,
244
+ "self": 0.0007145479999621784,
245
+ "children": {
246
+ "UnityEnvironment._generate_step_input": {
247
+ "total": 0.00034325699999726567,
248
+ "count": 1,
249
+ "is_parallel": true,
250
+ "self": 0.00034325699999726567
251
+ },
252
+ "communicator.exchange": {
253
+ "total": 0.036293932000035056,
254
+ "count": 1,
255
+ "is_parallel": true,
256
+ "self": 0.036293932000035056
257
+ },
258
+ "steps_from_proto": {
259
+ "total": 0.002094241999998303,
260
+ "count": 1,
261
+ "is_parallel": true,
262
+ "self": 0.000490292999927533,
263
+ "children": {
264
+ "_process_rank_one_or_two_observation": {
265
+ "total": 0.0016039490000707701,
266
+ "count": 10,
267
+ "is_parallel": true,
268
+ "self": 0.0016039490000707701
269
+ }
270
+ }
271
+ }
272
+ }
273
+ }
274
+ }
275
+ },
276
+ "UnityEnvironment.step": {
277
+ "total": 246.2607620609935,
278
+ "count": 18218,
279
+ "is_parallel": true,
280
+ "self": 9.9284912779936,
281
+ "children": {
282
+ "UnityEnvironment._generate_step_input": {
283
+ "total": 5.360988206002162,
284
+ "count": 18218,
285
+ "is_parallel": true,
286
+ "self": 5.360988206002162
287
+ },
288
+ "communicator.exchange": {
289
+ "total": 198.92844459200268,
290
+ "count": 18218,
291
+ "is_parallel": true,
292
+ "self": 198.92844459200268
293
+ },
294
+ "steps_from_proto": {
295
+ "total": 32.04283798499506,
296
+ "count": 18218,
297
+ "is_parallel": true,
298
+ "self": 7.090635067996004,
299
+ "children": {
300
+ "_process_rank_one_or_two_observation": {
301
+ "total": 24.952202916999056,
302
+ "count": 182180,
303
+ "is_parallel": true,
304
+ "self": 24.952202916999056
305
+ }
306
+ }
307
+ }
308
+ }
309
+ }
310
+ }
311
+ }
312
+ }
313
+ }
314
+ }
315
+ }
316
+ }
317
+ },
318
+ "trainer_threads": {
319
+ "total": 0.00019136000003072695,
320
+ "count": 1,
321
+ "self": 0.00019136000003072695,
322
+ "children": {
323
+ "thread_root": {
324
+ "total": 0.0,
325
+ "count": 0,
326
+ "is_parallel": true,
327
+ "self": 0.0,
328
+ "children": {
329
+ "trainer_advance": {
330
+ "total": 463.78587482701107,
331
+ "count": 408862,
332
+ "is_parallel": true,
333
+ "self": 10.26670675705094,
334
+ "children": {
335
+ "process_trajectory": {
336
+ "total": 264.46009117695996,
337
+ "count": 408862,
338
+ "is_parallel": true,
339
+ "self": 263.43942993495995,
340
+ "children": {
341
+ "RLTrainer._checkpoint": {
342
+ "total": 1.020661242000017,
343
+ "count": 4,
344
+ "is_parallel": true,
345
+ "self": 1.020661242000017
346
+ }
347
+ }
348
+ },
349
+ "_update_policy": {
350
+ "total": 189.05907689300017,
351
+ "count": 90,
352
+ "is_parallel": true,
353
+ "self": 68.96119403100369,
354
+ "children": {
355
+ "TorchPPOOptimizer.update": {
356
+ "total": 120.09788286199648,
357
+ "count": 4587,
358
+ "is_parallel": true,
359
+ "self": 120.09788286199648
360
+ }
361
+ }
362
+ }
363
+ }
364
+ }
365
+ }
366
+ }
367
+ }
368
+ },
369
+ "TrainerController._save_models": {
370
+ "total": 0.18918893900001876,
371
+ "count": 1,
372
+ "self": 0.0012132519999568103,
373
+ "children": {
374
+ "RLTrainer._checkpoint": {
375
+ "total": 0.18797568700006195,
376
+ "count": 1,
377
+ "self": 0.18797568700006195
378
+ }
379
+ }
380
+ }
381
+ }
382
+ }
383
+ }
384
+ }
run_logs/training_status.json ADDED
@@ -0,0 +1,65 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "SnowballTarget": {
3
+ "checkpoints": [
4
+ {
5
+ "steps": 49936,
6
+ "file_path": "results/SnowballTarget1/SnowballTarget/SnowballTarget-49936.onnx",
7
+ "reward": 13.727272727272727,
8
+ "creation_time": 1678019951.1009746,
9
+ "auxillary_file_paths": [
10
+ "results/SnowballTarget1/SnowballTarget/SnowballTarget-49936.pt"
11
+ ]
12
+ },
13
+ {
14
+ "steps": 99960,
15
+ "file_path": "results/SnowballTarget1/SnowballTarget/SnowballTarget-99960.onnx",
16
+ "reward": 19.545454545454547,
17
+ "creation_time": 1678020067.8229847,
18
+ "auxillary_file_paths": [
19
+ "results/SnowballTarget1/SnowballTarget/SnowballTarget-99960.pt"
20
+ ]
21
+ },
22
+ {
23
+ "steps": 149984,
24
+ "file_path": "results/SnowballTarget1/SnowballTarget/SnowballTarget-149984.onnx",
25
+ "reward": 24.727272727272727,
26
+ "creation_time": 1678020182.0007224,
27
+ "auxillary_file_paths": [
28
+ "results/SnowballTarget1/SnowballTarget/SnowballTarget-149984.pt"
29
+ ]
30
+ },
31
+ {
32
+ "steps": 199984,
33
+ "file_path": "results/SnowballTarget1/SnowballTarget/SnowballTarget-199984.onnx",
34
+ "reward": 26.363636363636363,
35
+ "creation_time": 1678020297.5775416,
36
+ "auxillary_file_paths": [
37
+ "results/SnowballTarget1/SnowballTarget/SnowballTarget-199984.pt"
38
+ ]
39
+ },
40
+ {
41
+ "steps": 200112,
42
+ "file_path": "results/SnowballTarget1/SnowballTarget/SnowballTarget-200112.onnx",
43
+ "reward": 26.363636363636363,
44
+ "creation_time": 1678020297.7918494,
45
+ "auxillary_file_paths": [
46
+ "results/SnowballTarget1/SnowballTarget/SnowballTarget-200112.pt"
47
+ ]
48
+ }
49
+ ],
50
+ "final_checkpoint": {
51
+ "steps": 200112,
52
+ "file_path": "results/SnowballTarget1/SnowballTarget.onnx",
53
+ "reward": 26.363636363636363,
54
+ "creation_time": 1678020297.7918494,
55
+ "auxillary_file_paths": [
56
+ "results/SnowballTarget1/SnowballTarget/SnowballTarget-200112.pt"
57
+ ]
58
+ }
59
+ },
60
+ "metadata": {
61
+ "stats_format_version": "0.3.0",
62
+ "mlagents_version": "0.29.0.dev0",
63
+ "torch_version": "1.8.1+cu102"
64
+ }
65
+ }