Trained with optuna hyperparam optimization with ml-agents-optuna

README.md

@@ -0,0 +1,35 @@
+---
+library_name: ml-agents
+tags:
+- SoccerTwos
+- deep-reinforcement-learning
+- reinforcement-learning
+- ML-Agents-SoccerTwos
+---
+
+  # **poca** Agent playing **SoccerTwos**
+  This is a trained model of a **poca** agent playing **SoccerTwos**
+  using the [Unity ML-Agents Library](https://github.com/Unity-Technologies/ml-agents).
+
+  ## Usage (with ML-Agents)
+  The Documentation: https://unity-technologies.github.io/ml-agents/ML-Agents-Toolkit-Documentation/
+
+  We wrote a complete tutorial to learn to train your first agent using ML-Agents and publish it to the Hub:
+  - A *short tutorial* where you teach Huggy the Dog 🐶 to fetch the stick and then play with him directly in your
+  browser: https://huggingface.co/learn/deep-rl-course/unitbonus1/introduction
+  - A *longer tutorial* to understand how works ML-Agents:
+  https://huggingface.co/learn/deep-rl-course/unit5/introduction
+
+  ### Resume the training
+  ```bash
+  mlagents-learn <your_configuration_file_path.yaml> --run-id=<run_id> --resume
+  ```
+
+  ### Watch your Agent play
+  You can watch your agent **playing directly in your browser**
+
+  1. If the environment is part of ML-Agents official environments, go to https://huggingface.co/unity
+  2. Step 1: Find your model_id: Statos6/poca-SoccerTwos
+  3. Step 2: Select your *.nn /*.onnx file
+  4. Click on Watch the agent play 👀
+  

SoccerTwos.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c574606e5c1839061cdb7dc9b76863444295b5b91caccfc413a1ed077ee8c3f
+size 895526

SoccerTwos/SoccerTwos-4999994.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:be96c3872475cc25d67472685bc2fee156427488a3c08e79bd28cf6c98213a94
+size 895526

SoccerTwos/SoccerTwos-4999994.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:71e54f36b571a60d1bd3474e5ee403ddc02aa6aafcd1c007e1272b0fa33cbfd8
+size 10325138

SoccerTwos/SoccerTwos-5000026.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:23416b67a78fe350c105debeb7be130b546d1ca99633a6d58b7b2ab627bdac15
+size 895526

SoccerTwos/SoccerTwos-5000026.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ef0aa7a1f48d9f4164990361bc247cd36df0a533121533f0cdf22a25adfb7189
+size 10325138

SoccerTwos/SoccerTwos-5499990.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5ee5c6c1bdaa2e03dd5acac35279c5a077f98827615be82d01ca79d646f2847f
+size 895526

SoccerTwos/SoccerTwos-5499990.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b1b631e6e474edfbd6397a6a173d3a36a447960b4e996fe4f72efd68936a6196
+size 10325138

SoccerTwos/SoccerTwos-5999976.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:56acb86e98bacbe50d2c8305c10c4bcb279131ff4c1359ef6245cccfed56cd97
+size 895526

SoccerTwos/SoccerTwos-5999976.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c17aec0f05d051f9f1c6ca2db48529936e5c49c6978ab68abb2f1a5a9dbc004a
+size 10325138

SoccerTwos/SoccerTwos-6000104.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c574606e5c1839061cdb7dc9b76863444295b5b91caccfc413a1ed077ee8c3f
+size 895526

SoccerTwos/SoccerTwos-6000104.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ccaa39ff525af5de7c6578077964314fe8b52d503f9fd6cb8fa52b144355b61b
+size 10325138

SoccerTwos/checkpoint.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:008145e1e808608cc93b4d650ceaa479d23cc271324dee6c3b01906b96fa84f6
+size 10323922

SoccerTwos/events.out.tfevents.1710290777.stephan-ms7c02.14283.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:921364813e102c784e2b1fe10a218e4108aaf0c18241e88faad09085f782460c
+size 4854988

SoccerTwos/events.out.tfevents.1710333085.stephan-ms7c02.24579.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7cd4e31cb3abbb0f4426bddd18591f073816db885a8bc2560ff7951c70d47284
+size 1017522

config.json

@@ -0,0 +1 @@
+{"default_settings": null, "behaviors": {"SoccerTwos": {"trainer_type": "poca", "hyperparameters": {"batch_size": 4096, "buffer_size": 65536, "learning_rate": 0.0004345843188705317, "beta": 0.00019215097636640332, "epsilon": 0.15831928707038231, "lambd": 0.9467147291117369, "num_epoch": 5, "learning_rate_schedule": "linear", "beta_schedule": "linear", "epsilon_schedule": "linear"}, "checkpoint_interval": 500000, "network_settings": {"normalize": true, "hidden_units": 256, "num_layers": 3, "vis_encode_type": "fully_connected", "memory": null, "goal_conditioning_type": "hyper", "deterministic": false}, "reward_signals": {"extrinsic": {"gamma": 0.9608227414353802, "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": 5, "even_checkpoints": false, "max_steps": 6000000, "time_horizon": 32, "summary_freq": 10000, "threaded": false, "self_play": {"save_steps": 60000, "team_change": 589998, "swap_steps": 70000, "window": 25, "play_against_latest_model_ratio": 0.2427456302841774, "initial_elo": 1200.0}, "behavioral_cloning": null}}, "env_settings": {"env_path": "./ml-agents/training-envs-executables/SoccerTwos.exe", "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": 1280, "height": 720, "quality_level": 5, "time_scale": 20, "target_frame_rate": -1, "capture_frame_rate": 60, "no_graphics": false, "no_graphics_monitor": false}, "environment_parameters": null, "checkpoint_settings": {"run_id": "SoccerTwos-read-deal", "initialize_from": null, "load_model": false, "resume": true, "force": false, "train_model": false, "inference": false, "results_dir": "results"}, "torch_settings": {"device": null}, "debug": false}

configuration.yaml

@@ -0,0 +1,84 @@
+default_settings: null
+behaviors:
+  SoccerTwos:
+    trainer_type: poca
+    hyperparameters:
+      batch_size: 4096
+      buffer_size: 65536
+      learning_rate: 0.0004345843188705317
+      beta: 0.00019215097636640332
+      epsilon: 0.15831928707038231
+      lambd: 0.9467147291117369
+      num_epoch: 5
+      learning_rate_schedule: linear
+      beta_schedule: linear
+      epsilon_schedule: linear
+    checkpoint_interval: 500000
+    network_settings:
+      normalize: true
+      hidden_units: 256
+      num_layers: 3
+      vis_encode_type: fully_connected
+      memory: null
+      goal_conditioning_type: hyper
+      deterministic: false
+    reward_signals:
+      extrinsic:
+        gamma: 0.9608227414353802
+        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: 5
+    even_checkpoints: false
+    max_steps: 6000000
+    time_horizon: 32
+    summary_freq: 10000
+    threaded: false
+    self_play:
+      save_steps: 60000
+      team_change: 589998
+      swap_steps: 70000
+      window: 25
+      play_against_latest_model_ratio: 0.2427456302841774
+      initial_elo: 1200.0
+    behavioral_cloning: null
+env_settings:
+  env_path: ./ml-agents/training-envs-executables/SoccerTwos.exe
+  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: 1280
+  height: 720
+  quality_level: 5
+  time_scale: 20
+  target_frame_rate: -1
+  capture_frame_rate: 60
+  no_graphics: false
+  no_graphics_monitor: false
+environment_parameters: null
+checkpoint_settings:
+  run_id: SoccerTwos-read-deal
+  initialize_from: null
+  load_model: false
+  resume: true
+  force: false
+  train_model: false
+  inference: false
+  results_dir: results
+torch_settings:
+  device: null
+debug: false

run_logs/Player-0.log

@@ -0,0 +1,300 @@
+Mono path[0] = '/run/media/stephan/Linux_Secondary/RL/soccer/ml-agents/training-envs-executables/SoccerTwos_Data/Managed'
+Mono config path = '/run/media/stephan/Linux_Secondary/RL/soccer/ml-agents/training-envs-executables/SoccerTwos_Data/MonoBleedingEdge/etc'
+Preloaded 'lib_burst_generated.so'
+Preloaded 'libgrpc_csharp_ext.x64.so'
+Display 0 'DP-3 27"': 2560x1440 (primary device).
+Display 1 'HDMI-A-1 27"': 1920x1080 (secondary device).
+Desktop is 1920 x 1080 @ 60 Hz
+Initialize engine version: 2021.3.14f1 (eee1884e7226)
+[Subsystems] Discovering subsystems at path /run/media/stephan/Linux_Secondary/RL/soccer/ml-agents/training-envs-executables/SoccerTwos_Data/UnitySubsystems
+GfxDevice: creating device client; threaded=1; jobified=0
+Renderer: AMD Radeon RX Vega (radeonsi, vega10, LLVM 16.0.6, DRM 3.56, 6.7.0-1-rt6-MANJARO)
+Vendor:   AMD
+Version:  4.6 (Core Profile) Mesa 24.0.2-manjaro1.1
+GLES:     0
+ GL_AMD_conservative_depth GL_AMD_depth_clamp_separate GL_AMD_draw_buffers_blend GL_AMD_framebuffer_multisample_advanced GL_AMD_gpu_shader_int64 GL_AMD_multi_draw_indirect GL_AMD_performance_monitor GL_AMD_pinned_memory GL_AMD_query_buffer_object GL_AMD_seamless_cubemap_per_texture GL_AMD_shader_stencil_export GL_AMD_shader_trinary_minmax GL_AMD_texture_texture4 GL_AMD_vertex_shader_layer GL_AMD_vertex_shader_viewport_index GL_ANGLE_texture_compression_dxt3 GL_ANGLE_texture_compression_dxt5 GL_ARB_ES2_compatibility GL_ARB_ES3_1_compatibility GL_ARB_ES3_2_compatibility GL_ARB_ES3_compatibility GL_ARB_arrays_of_arrays GL_ARB_base_instance GL_ARB_bindless_texture GL_ARB_blend_func_extended GL_ARB_buffer_storage GL_ARB_clear_buffer_object GL_ARB_clear_texture GL_ARB_clip_control GL_ARB_color_buffer_float GL_ARB_compressed_texture_pixel_storage GL_ARB_compute_shader GL_ARB_compute_variable_group_size GL_ARB_conditional_render_inverted GL_ARB_conservative_depth GL_ARB_copy_buffer GL_ARB_copy_image GL_ARB_cull_dist
+ance GL_ARB_debug_output GL_ARB_depth_buffer_float GL_ARB_depth_clamp GL_ARB_derivative_control GL_ARB_direct_state_access GL_ARB_draw_buffers GL_ARB_draw_buffers_blend GL_ARB_draw_elements_base_vertex GL_ARB_draw_indirect GL_ARB_draw_instanced GL_ARB_enhanced_layouts GL_ARB_explicit_attrib_location GL_ARB_explicit_uniform_location GL_ARB_fragment_coord_conventions GL_ARB_fragment_layer_viewport GL_ARB_fragment_shader GL_ARB_framebuffer_no_attachments GL_ARB_framebuffer_object GL_ARB_framebuffer_sRGB GL_ARB_get_program_binary GL_ARB_get_texture_sub_image GL_ARB_gl_spirv GL_ARB_gpu_shader5 GL_ARB_gpu_shader_fp64 GL_ARB_gpu_shader_int64 GL_ARB_half_float_pixel GL_ARB_half_float_vertex GL_ARB_indirect_parameters GL_ARB_instanced_arrays GL_ARB_internalformat_query GL_ARB_internalformat_query2 GL_ARB_invalidate_subdata GL_ARB_map_buffer_alignment GL_ARB_map_buffer_range GL_ARB_multi_bind GL_ARB_multi_draw_indirect GL_ARB_occlusion_query2 GL_ARB_parallel_shader_compile GL_ARB_pipeline_statistics_query GL_ARB_pixel
+_buffer_object GL_ARB_point_sprite GL_ARB_polygon_offset_clamp GL_ARB_program_interface_query GL_ARB_provoking_vertex GL_ARB_query_buffer_object GL_ARB_robust_buffer_access_behavior GL_ARB_robustness GL_ARB_sample_shading GL_ARB_sampler_objects GL_ARB_seamless_cube_map GL_ARB_seamless_cubemap_per_texture GL_ARB_separate_shader_objects GL_ARB_shader_atomic_counter_ops GL_ARB_shader_atomic_counters GL_ARB_shader_ballot GL_ARB_shader_bit_encoding GL_ARB_shader_clock GL_ARB_shader_draw_parameters GL_ARB_shader_group_vote GL_ARB_shader_image_load_store GL_ARB_shader_image_size GL_ARB_shader_objects GL_ARB_shader_precision GL_ARB_shader_stencil_export GL_ARB_shader_storage_buffer_object GL_ARB_shader_subroutine GL_ARB_shader_texture_image_samples GL_ARB_shader_texture_lod GL_ARB_shader_viewport_layer_array GL_ARB_shading_language_420pack GL_ARB_shading_language_include GL_ARB_shading_language_packing GL_ARB_sparse_buffer GL_ARB_sparse_texture GL_ARB_sparse_texture2 GL_ARB_sparse_texture_clamp GL_ARB_spirv_extensio
+ns GL_ARB_stencil_texturing GL_ARB_sync GL_ARB_tessellation_shader GL_ARB_texture_barrier GL_ARB_texture_border_clamp GL_ARB_texture_buffer_object GL_ARB_texture_buffer_object_rgb32 GL_ARB_texture_buffer_range GL_ARB_texture_compression_bptc GL_ARB_texture_compression_rgtc GL_ARB_texture_cube_map_array GL_ARB_texture_filter_anisotropic GL_ARB_texture_float GL_ARB_texture_gather GL_ARB_texture_mirror_clamp_to_edge GL_ARB_texture_multisample GL_ARB_texture_non_power_of_two GL_ARB_texture_query_levels GL_ARB_texture_query_lod GL_ARB_texture_rectangle GL_ARB_texture_rg GL_ARB_texture_rgb10_a2ui GL_ARB_texture_stencil8 GL_ARB_texture_storage GL_ARB_texture_storage_multisample GL_ARB_texture_swizzle GL_ARB_texture_view GL_ARB_timer_query GL_ARB_transform_feedback2 GL_ARB_transform_feedback3 GL_ARB_transform_feedback_instanced GL_ARB_transform_feedback_overflow_query GL_ARB_uniform_buffer_object GL_ARB_vertex_array_bgra GL_ARB_vertex_array_object GL_ARB_vertex_attrib_64bit GL_ARB_vertex_attrib_binding GL_ARB_vertex
+_buffer_object GL_ARB_vertex_shader GL_ARB_vertex_type_10f_11f_11f_rev GL_ARB_vertex_type_2_10_10_10_rev GL_ARB_viewport_array GL_ATI_blend_equation_separate GL_ATI_meminfo GL_ATI_texture_float GL_ATI_texture_mirror_once GL_EXT_EGL_image_storage GL_EXT_EGL_sync GL_EXT_abgr GL_EXT_blend_equation_separate GL_EXT_debug_label GL_EXT_demote_to_helper_invocation GL_EXT_depth_bounds_test GL_EXT_draw_buffers2 GL_EXT_draw_instanced GL_EXT_framebuffer_blit GL_EXT_framebuffer_multisample GL_EXT_framebuffer_multisample_blit_scaled GL_EXT_framebuffer_object GL_EXT_framebuffer_sRGB GL_EXT_memory_object GL_EXT_memory_object_fd GL_EXT_packed_depth_stencil GL_EXT_packed_float GL_EXT_pixel_buffer_object GL_EXT_polygon_offset_clamp GL_EXT_provoking_vertex GL_EXT_semaphore GL_EXT_semaphore_fd GL_EXT_shader_image_load_formatted GL_EXT_shader_image_load_store GL_EXT_shader_integer_mix GL_EXT_shader_samples_identical GL_EXT_texture_array GL_EXT_texture_compression_dxt1 GL_EXT_texture_compression_rgtc GL_EXT_texture_compression_s3t
+c GL_EXT_texture_filter_anisotropic GL_EXT_texture_integer GL_EXT_texture_mirror_clamp GL_EXT_texture_sRGB GL_EXT_texture_sRGB_R8 GL_EXT_texture_sRGB_decode GL_EXT_texture_shadow_lod GL_EXT_texture_shared_exponent GL_EXT_texture_snorm GL_EXT_texture_swizzle GL_EXT_timer_query GL_EXT_transform_feedback GL_EXT_vertex_array_bgra GL_EXT_vertex_attrib_64bit GL_EXT_window_rectangles GL_IBM_multimode_draw_arrays GL_INTEL_blackhole_render GL_KHR_blend_equation_advanced GL_KHR_context_flush_control GL_KHR_debug GL_KHR_no_error GL_KHR_parallel_shader_compile GL_KHR_robust_buffer_access_behavior GL_KHR_robustness GL_KHR_texture_compression_astc_ldr GL_KHR_texture_compression_astc_sliced_3d GL_MESA_framebuffer_flip_y GL_MESA_pack_invert GL_MESA_shader_integer_functions GL_MESA_texture_const_bandwidth GL_MESA_texture_signed_rgba GL_NVX_gpu_memory_info GL_NV_alpha_to_coverage_dither_control GL_NV_compute_shader_derivatives GL_NV_conditional_render GL_NV_copy_image GL_NV_depth_clamp GL_NV_packed_depth_stencil GL_NV_shader_
+atomic_int64 GL_NV_texture_barrier GL_NV_vdpau_interop GL_OES_EGL_image GL_S3_s3tc
+OPENGL LOG: Creating OpenGL 4.6 graphics device ; Context level  <OpenGL 4.5> ; Context handle 16352240
+Begin MonoManager ReloadAssembly
+- Completed reload, in  0.083 seconds
+Default vsync count 1
+requesting resize 1280 x 720
+resizing window to 1280 x 720
+Desktop is 1920 x 1080 @ 60 Hz
+UnloadTime: 0.827430 ms
+Registered Communicator in Agent.
+Setting up 6 worker threads for Enlighten.
+requesting resize 1280 x 720
+resizing window to 1280 x 720
+Desktop is 1920 x 1080 @ 60 Hz
+Memory Statistics:
+[ALLOC_TEMP_TLS] TLS Allocator
+  StackAllocators : 
+    [ALLOC_TEMP_MAIN]
+      Peak usage frame count: [8.0 KB-16.0 KB]: 18758 frames, [2.0 MB-4.0 MB]: 1 frames
+      Initial Block Size 4.0 MB
+      Current Block Size 4.0 MB
+      Peak Allocated Bytes 2.1 MB
+      Overflow Count 0
+    [ALLOC_TEMP_Loading.AsyncRead]
+      Initial Block Size 64.0 KB
+      Current Block Size 64.0 KB
+      Peak Allocated Bytes 128 B
+      Overflow Count 0
+    [ALLOC_TEMP_Loading.PreloadManager]
+      Initial Block Size 256.0 KB
+      Current Block Size 300.0 KB
+      Peak Allocated Bytes 249.6 KB
+      Overflow Count 4
+    [ALLOC_TEMP_Background Job.Worker 8]
+      Initial Block Size 32.0 KB
+      Current Block Size 32.0 KB
+      Peak Allocated Bytes 0 B
+      Overflow Count 0
+    [ALLOC_TEMP_Job.Worker 6]
+      Initial Block Size 256.0 KB
+      Current Block Size 256.0 KB
+      Peak Allocated Bytes 1.5 KB
+      Overflow Count 0
+    [ALLOC_TEMP_Job.Worker 0]
+      Initial Block Size 256.0 KB
+      Current Block Size 256.0 KB
+      Peak Allocated Bytes 1.5 KB
+      Overflow Count 0
+    [ALLOC_TEMP_Background Job.Worker 10]
+      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 9]
+      Initial Block Size 32.0 KB
+      Current Block Size 32.0 KB
+      Peak Allocated Bytes 0 B
+      Overflow Count 0
+    [ALLOC_TEMP_Job.Worker 5]
+      Initial Block Size 256.0 KB
+      Current Block Size 256.0 KB
+      Peak Allocated Bytes 1.5 KB
+      Overflow Count 0
+    [ALLOC_TEMP_Background Job.Worker 14]
+      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 6]
+      Initial Block Size 32.0 KB
+      Current Block Size 32.0 KB
+      Peak Allocated Bytes 0 B
+      Overflow Count 0
+    [ALLOC_TEMP_Job.Worker 4]
+      Initial Block Size 256.0 KB
+      Current Block Size 256.0 KB
+      Peak Allocated Bytes 1.5 KB
+      Overflow Count 0
+    [ALLOC_TEMP_Job.Worker 3]
+      Initial Block Size 256.0 KB
+      Current Block Size 256.0 KB
+      Peak Allocated Bytes 1.5 KB
+      Overflow Count 0
+    [ALLOC_TEMP_Background Job.Worker 12]
+      Initial Block Size 32.0 KB
+      Current Block Size 32.0 KB
+      Peak Allocated Bytes 0 B
+      Overflow Count 0
+    [ALLOC_TEMP_EnlightenWorker] x 6
+      Initial Block Size 64.0 KB
+      Current Block Size 64.0 KB
+      Peak Allocated Bytes 0 B
+      Overflow Count 0
+    [ALLOC_TEMP_Background Job.Worker 15]
+      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 1]
+      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]
+      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]
+      Initial Block Size 32.0 KB
+      Current Block Size 32.0 KB
+      Peak Allocated Bytes 0 B
+      Overflow Count 0
+    [ALLOC_TEMP_UnityGfxDeviceWorker]
+      Initial Block Size 256.0 KB
+      Current Block Size 256.0 KB
+      Peak Allocated Bytes 45.5 KB
+      Overflow Count 0
+    [ALLOC_TEMP_AssetGarbageCollectorHelper] x 11
+      Initial Block Size 64.0 KB
+      Current Block Size 64.0 KB
+      Peak Allocated Bytes 0 B
+      Overflow Count 0
+    [ALLOC_TEMP_Background Job.Worker 5]
+      Initial Block Size 32.0 KB
+      Current Block Size 32.0 KB
+      Peak Allocated Bytes 0 B
+      Overflow Count 0
+    [ALLOC_TEMP_Job.Worker 9]
+      Initial Block Size 256.0 KB
+      Current Block Size 256.0 KB
+      Peak Allocated Bytes 1.5 KB
+      Overflow Count 0
+    [ALLOC_TEMP_Background Job.Worker 13]
+      Initial Block Size 32.0 KB
+      Current Block Size 32.0 KB
+      Peak Allocated Bytes 0 B
+      Overflow Count 0
+    [ALLOC_TEMP_Job.Worker 8]
+      Initial Block Size 256.0 KB
+      Current Block Size 256.0 KB
+      Peak Allocated Bytes 1.5 KB
+      Overflow Count 0
+    [ALLOC_TEMP_Job.Worker 7]
+      Initial Block Size 256.0 KB
+      Current Block Size 256.0 KB
+      Peak Allocated Bytes 1.5 KB
+      Overflow Count 0
+    [ALLOC_TEMP_Job.Worker 1]
+      Initial Block Size 256.0 KB
+      Current Block Size 256.0 KB
+      Peak Allocated Bytes 1.5 KB
+      Overflow Count 0
+    [ALLOC_TEMP_Job.Worker 2]
+      Initial Block Size 256.0 KB
+      Current Block Size 256.0 KB
+      Peak Allocated Bytes 1.5 KB
+      Overflow Count 0
+    [ALLOC_TEMP_Background Job.Worker 3]
+      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 11]
+      Initial Block Size 32.0 KB
+      Current Block Size 32.0 KB
+      Peak Allocated Bytes 0 B
+      Overflow Count 0
+    [ALLOC_TEMP_Job.Worker 10]
+      Initial Block Size 256.0 KB
+      Current Block Size 256.0 KB
+      Peak Allocated Bytes 1.5 KB
+      Overflow Count 0
+    [ALLOC_TEMP_Background Job.Worker 0]
+      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 4]
+      Initial Block Size 32.0 KB
+      Current Block Size 32.0 KB
+      Peak Allocated Bytes 0 B
+      Overflow Count 0
+    [ALLOC_TEMP_BatchDeleteObjects]
+      Initial Block Size 64.0 KB
+      Current Block Size 64.0 KB
+      Peak Allocated Bytes 0 B
+      Overflow Count 0
+[ALLOC_DEFAULT] Dual Thread Allocator
+  Peak main deferred allocation count 27
+    [ALLOC_BUCKET]
+      Large Block size 4.0 MB
+      Used Block count 1
+      Peak Allocated bytes 2.0 MB
+    [ALLOC_DEFAULT_MAIN]
+      Peak usage frame count: [8.0 MB-16.0 MB]: 18759 frames
+      Requested Block Size 16.0 MB
+      Peak Block count 1
+      Peak Allocated memory 13.0 MB
+      Peak Large allocation bytes 0 B
+    [ALLOC_DEFAULT_THREAD]
+      Peak usage frame count: [16.0 MB-32.0 MB]: 18759 frames
+      Requested Block Size 16.0 MB
+      Peak Block count 1
+      Peak Allocated memory 29.1 MB
+      Peak Large allocation bytes 16.0 MB
+[ALLOC_TEMP_JOB_1_FRAME]
+  Initial Block Size 2.0 MB
+  Used Block Count 1
+  Overflow Count (too large) 0
+  Overflow Count (full) 0
+[ALLOC_TEMP_JOB_2_FRAMES]
+  Initial Block Size 2.0 MB
+  Used Block Count 1
+  Overflow Count (too large) 0
+  Overflow Count (full) 0
+[ALLOC_TEMP_JOB_4_FRAMES (JobTemp)]
+  Initial Block Size 2.0 MB
+  Used Block Count 2
+  Overflow Count (too large) 0
+  Overflow Count (full) 0
+[ALLOC_TEMP_JOB_ASYNC (Background)]
+  Initial Block Size 1.0 MB
+  Used Block Count 3
+  Overflow Count (too large) 0
+  Overflow Count (full) 0
+[ALLOC_GFX] Dual Thread Allocator
+  Peak main deferred allocation count 1
+    [ALLOC_BUCKET]
+      Large Block size 4.0 MB
+      Used Block count 1
+      Peak Allocated bytes 2.0 MB
+    [ALLOC_GFX_MAIN]
+      Peak usage frame count: [32.0 KB-64.0 KB]: 18758 frames, [2.0 MB-4.0 MB]: 1 frames
+      Requested Block Size 16.0 MB
+      Peak Block count 1
+      Peak Allocated memory 2.7 MB
+      Peak Large allocation bytes 0 B
+    [ALLOC_GFX_THREAD]
+      Peak usage frame count: [128.0 KB-256.0 KB]: 18759 frames
+      Requested Block Size 16.0 MB
+      Peak Block count 1
+      Peak Allocated memory 174.7 KB
+      Peak Large allocation bytes 0 B
+[ALLOC_CACHEOBJECTS] Dual Thread Allocator
+  Peak main deferred allocation count 0
+    [ALLOC_BUCKET]
+      Large Block size 4.0 MB
+      Used Block count 1
+      Peak Allocated bytes 2.0 MB
+    [ALLOC_CACHEOBJECTS_MAIN]
+      Peak usage frame count: [0.5 MB-1.0 MB]: 18758 frames, [1.0 MB-2.0 MB]: 1 frames
+      Requested Block Size 4.0 MB
+      Peak Block count 1
+      Peak Allocated memory 1.2 MB
+      Peak Large allocation bytes 0 B
+    [ALLOC_CACHEOBJECTS_THREAD]
+      Peak usage frame count: [4.0 MB-8.0 MB]: 18759 frames
+      Requested Block Size 4.0 MB
+      Peak Block count 2
+      Peak Allocated memory 7.8 MB
+      Peak Large allocation bytes 0 B
+[ALLOC_TYPETREE] Dual Thread Allocator
+  Peak main deferred allocation count 0
+    [ALLOC_BUCKET]
+      Large Block size 4.0 MB
+      Used Block count 1
+      Peak Allocated bytes 2.0 MB
+    [ALLOC_TYPETREE_MAIN]
+      Peak usage frame count: [0-1.0 KB]: 18759 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]: 18758 frames, [2.0 KB-4.0 KB]: 1 frames
+      Requested Block Size 2.0 MB
+      Peak Block count 1
+      Peak Allocated memory 2.2 KB
+      Peak Large allocation bytes 0 B

run_logs/timers.json

@@ -0,0 +1,424 @@
+{
+    "name": "root",
+    "gauges": {
+        "SoccerTwos.Policy.Entropy.mean": {
+            "value": 1.778828501701355,
+            "min": 1.7496646642684937,
+            "max": 1.904069185256958,
+            "count": 100
+        },
+        "SoccerTwos.Policy.Entropy.sum": {
+            "value": 35690.4140625,
+            "min": 34937.3046875,
+            "max": 38751.6171875,
+            "count": 100
+        },
+        "SoccerTwos.Environment.EpisodeLength.mean": {
+            "value": 43.51754385964912,
+            "min": 38.792,
+            "max": 53.82022471910113,
+            "count": 100
+        },
+        "SoccerTwos.Environment.EpisodeLength.sum": {
+            "value": 19844.0,
+            "min": 18336.0,
+            "max": 20936.0,
+            "count": 100
+        },
+        "SoccerTwos.Self-play.ELO.mean": {
+            "value": 1561.460064375993,
+            "min": 1519.7251303553944,
+            "max": 1561.4942652168481,
+            "count": 100
+        },
+        "SoccerTwos.Self-play.ELO.sum": {
+            "value": 356012.8946777264,
+            "min": 273814.30154407705,
+            "max": 390373.566304212,
+            "count": 100
+        },
+        "SoccerTwos.Step.mean": {
+            "value": 5999976.0,
+            "min": 5009984.0,
+            "max": 5999976.0,
+            "count": 100
+        },
+        "SoccerTwos.Step.sum": {
+            "value": 5999976.0,
+            "min": 5009984.0,
+            "max": 5999976.0,
+            "count": 100
+        },
+        "SoccerTwos.Policy.ExtrinsicBaselineEstimate.mean": {
+            "value": 0.036884572356939316,
+            "min": -0.06633567065000534,
+            "max": 0.09457974880933762,
+            "count": 100
+        },
+        "SoccerTwos.Policy.ExtrinsicBaselineEstimate.sum": {
+            "value": 15.380867004394531,
+            "min": -27.06495475769043,
+            "max": 38.872276306152344,
+            "count": 100
+        },
+        "SoccerTwos.Policy.ExtrinsicValueEstimate.mean": {
+            "value": 0.03724776208400726,
+            "min": -0.06787735223770142,
+            "max": 0.09511835128068924,
+            "count": 100
+        },
+        "SoccerTwos.Policy.ExtrinsicValueEstimate.sum": {
+            "value": 15.532317161560059,
+            "min": -27.693960189819336,
+            "max": 39.09364318847656,
+            "count": 100
+        },
+        "SoccerTwos.Environment.CumulativeReward.mean": {
+            "value": 0.0,
+            "min": 0.0,
+            "max": 0.0,
+            "count": 100
+        },
+        "SoccerTwos.Environment.CumulativeReward.sum": {
+            "value": 0.0,
+            "min": 0.0,
+            "max": 0.0,
+            "count": 100
+        },
+        "SoccerTwos.Policy.ExtrinsicReward.mean": {
+            "value": 0.04650350934580753,
+            "min": -0.2562914263634455,
+            "max": 0.2151981968063492,
+            "count": 100
+        },
+        "SoccerTwos.Policy.ExtrinsicReward.sum": {
+            "value": 10.602800130844116,
+            "min": -53.82119953632355,
+            "max": 47.77399969100952,
+            "count": 100
+        },
+        "SoccerTwos.Environment.GroupCumulativeReward.mean": {
+            "value": 0.04650350934580753,
+            "min": -0.2562914263634455,
+            "max": 0.2151981968063492,
+            "count": 100
+        },
+        "SoccerTwos.Environment.GroupCumulativeReward.sum": {
+            "value": 10.602800130844116,
+            "min": -53.82119953632355,
+            "max": 47.77399969100952,
+            "count": 100
+        },
+        "SoccerTwos.IsTraining.mean": {
+            "value": 1.0,
+            "min": 1.0,
+            "max": 1.0,
+            "count": 100
+        },
+        "SoccerTwos.IsTraining.sum": {
+            "value": 1.0,
+            "min": 1.0,
+            "max": 1.0,
+            "count": 100
+        },
+        "SoccerTwos.Losses.PolicyLoss.mean": {
+            "value": 0.014634914088446749,
+            "min": 0.010371605570253451,
+            "max": 0.014634914088446749,
+            "count": 15
+        },
+        "SoccerTwos.Losses.PolicyLoss.sum": {
+            "value": 0.014634914088446749,
+            "min": 0.010371605570253451,
+            "max": 0.014634914088446749,
+            "count": 15
+        },
+        "SoccerTwos.Losses.ValueLoss.mean": {
+            "value": 0.051349658705294135,
+            "min": 0.048586570238694546,
+            "max": 0.05451308977790177,
+            "count": 15
+        },
+        "SoccerTwos.Losses.ValueLoss.sum": {
+            "value": 0.051349658705294135,
+            "min": 0.048586570238694546,
+            "max": 0.05451308977790177,
+            "count": 15
+        },
+        "SoccerTwos.Losses.BaselineLoss.mean": {
+            "value": 0.0515909023117274,
+            "min": 0.04938719081692398,
+            "max": 0.054688979778438805,
+            "count": 15
+        },
+        "SoccerTwos.Losses.BaselineLoss.sum": {
+            "value": 0.0515909023117274,
+            "min": 0.04938719081692398,
+            "max": 0.054688979778438805,
+            "count": 15
+        },
+        "SoccerTwos.Policy.LearningRate.mean": {
+            "value": 1.1853560262662174e-06,
+            "min": 1.1853560262662174e-06,
+            "max": 6.767456859236319e-05,
+            "count": 15
+        },
+        "SoccerTwos.Policy.LearningRate.sum": {
+            "value": 1.1853560262662174e-06,
+            "min": 1.1853560262662174e-06,
+            "max": 6.767456859236319e-05,
+            "count": 15
+        },
+        "SoccerTwos.Policy.Epsilon.mean": {
+            "value": 0.1001590561356033,
+            "min": 0.1001590561356033,
+            "max": 0.10908161546093645,
+            "count": 15
+        },
+        "SoccerTwos.Policy.Epsilon.sum": {
+            "value": 0.1001590561356033,
+            "min": 0.1001590561356033,
+            "max": 0.10908161546093645,
+            "count": 15
+        },
+        "SoccerTwos.Policy.Beta.mean": {
+            "value": 1.04967864295433e-05,
+            "min": 1.04967864295433e-05,
+            "max": 3.8364975058721184e-05,
+            "count": 15
+        },
+        "SoccerTwos.Policy.Beta.sum": {
+            "value": 1.04967864295433e-05,
+            "min": 1.04967864295433e-05,
+            "max": 3.8364975058721184e-05,
+            "count": 15
+        }
+    },
+    "metadata": {
+        "timer_format_version": "0.1.0",
+        "start_time_seconds": "1710333081",
+        "python_version": "3.10.12 (main, Jul  5 2023, 18:54:27) [GCC 11.2.0]",
+        "command_line_arguments": "/home/stephan/.conda/envs/rl/bin/mlagents-learn ./ml-agents/config/poca/SoccerTwos.yaml --env=./ml-agents/training-envs-executables/SoccerTwos.exe --run-id=SoccerTwos-read-deal --width=1280 --height=720 --resume",
+        "mlagents_version": "1.1.0.dev0",
+        "mlagents_envs_version": "1.1.0.dev0",
+        "communication_protocol_version": "1.5.0",
+        "pytorch_version": "2.2.1+cu121",
+        "numpy_version": "1.23.5",
+        "end_time_seconds": "1710335865"
+    },
+    "total": 2783.08172296299,
+    "count": 1,
+    "self": 0.3221229049813701,
+    "children": {
+        "run_training.setup": {
+            "total": 0.017381374011165462,
+            "count": 1,
+            "self": 0.017381374011165462
+        },
+        "TrainerController.start_learning": {
+            "total": 2782.7422186839976,
+            "count": 1,
+            "self": 1.4529281045834068,
+            "children": {
+                "TrainerController._reset_env": {
+                    "total": 4.399289562992635,
+                    "count": 3,
+                    "self": 4.399289562992635
+                },
+                "TrainerController.advance": {
+                    "total": 2776.7682698464196,
+                    "count": 71154,
+                    "self": 1.2270896326372167,
+                    "children": {
+                        "env_step": {
+                            "total": 1269.4150066098955,
+                            "count": 71154,
+                            "self": 1077.1618315594678,
+                            "children": {
+                                "SubprocessEnvManager._take_step": {
+                                    "total": 191.40557155759598,
+                                    "count": 71154,
+                                    "self": 5.71247391810175,
+                                    "children": {
+                                        "TorchPolicy.evaluate": {
+                                            "total": 185.69309763949423,
+                                            "count": 125058,
+                                            "self": 185.69309763949423
+                                        }
+                                    }
+                                },
+                                "workers": {
+                                    "total": 0.8476034928316949,
+                                    "count": 71154,
+                                    "self": 0.0,
+                                    "children": {
+                                        "worker_root": {
+                                            "total": 2780.0714288110757,
+                                            "count": 71154,
+                                            "is_parallel": true,
+                                            "self": 1856.7726158039295,
+                                            "children": {
+                                                "run_training.setup": {
+                                                    "total": 0.0,
+                                                    "count": 0,
+                                                    "is_parallel": true,
+                                                    "self": 0.0,
+                                                    "children": {
+                                                        "steps_from_proto": {
+                                                            "total": 0.06613134100916795,
+                                                            "count": 2,
+                                                            "is_parallel": true,
+                                                            "self": 0.00047555101627949625,
+                                                            "children": {
+                                                                "_process_rank_one_or_two_observation": {
+                                                                    "total": 0.06565578999288846,
+                                                                    "count": 8,
+                                                                    "is_parallel": true,
+                                                                    "self": 0.06565578999288846
+                                                                }
+                                                            }
+                                                        },
+                                                        "UnityEnvironment.step": {
+                                                            "total": 0.031780133009306155,
+                                                            "count": 1,
+                                                            "is_parallel": true,
+                                                            "self": 0.0006475910049630329,
+                                                            "children": {
+                                                                "UnityEnvironment._generate_step_input": {
+                                                                    "total": 0.0005751600110670552,
+                                                                    "count": 1,
+                                                                    "is_parallel": true,
+                                                                    "self": 0.0005751600110670552
+                                                                },
+                                                                "communicator.exchange": {
+                                                                    "total": 0.02851476900104899,
+                                                                    "count": 1,
+                                                                    "is_parallel": true,
+                                                                    "self": 0.02851476900104899
+                                                                },
+                                                                "steps_from_proto": {
+                                                                    "total": 0.0020426129922270775,
+                                                                    "count": 2,
+                                                                    "is_parallel": true,
+                                                                    "self": 0.0004071999865118414,
+                                                                    "children": {
+                                                                        "_process_rank_one_or_two_observation": {
+                                                                            "total": 0.001635413005715236,
+                                                                            "count": 8,
+                                                                            "is_parallel": true,
+                                                                            "self": 0.001635413005715236
+                                                                        }
+                                                                    }
+                                                                }
+                                                            }
+                                                        }
+                                                    }
+                                                },
+                                                "steps_from_proto": {
+                                                    "total": 0.003298083000117913,
+                                                    "count": 4,
+                                                    "is_parallel": true,
+                                                    "self": 0.0005663799965986982,
+                                                    "children": {
+                                                        "_process_rank_one_or_two_observation": {
+                                                            "total": 0.0027317030035192147,
+                                                            "count": 16,
+                                                            "is_parallel": true,
+                                                            "self": 0.0027317030035192147
+                                                        }
+                                                    }
+                                                },
+                                                "UnityEnvironment.step": {
+                                                    "total": 923.2955149241461,
+                                                    "count": 71153,
+                                                    "is_parallel": true,
+                                                    "self": 50.94944198477606,
+                                                    "children": {
+                                                        "UnityEnvironment._generate_step_input": {
+                                                            "total": 31.86183314715163,
+                                                            "count": 71153,
+                                                            "is_parallel": true,
+                                                            "self": 31.86183314715163
+                                                        },
+                                                        "communicator.exchange": {
+                                                            "total": 713.6419240085379,
+                                                            "count": 71153,
+                                                            "is_parallel": true,
+                                                            "self": 713.6419240085379
+                                                        },
+                                                        "steps_from_proto": {
+                                                            "total": 126.84231578368053,
+                                                            "count": 142306,
+                                                            "is_parallel": true,
+                                                            "self": 22.54734104256204,
+                                                            "children": {
+                                                                "_process_rank_one_or_two_observation": {
+                                                                    "total": 104.2949747411185,
+                                                                    "count": 569224,
+                                                                    "is_parallel": true,
+                                                                    "self": 104.2949747411185
+                                                                }
+                                                            }
+                                                        }
+                                                    }
+                                                }
+                                            }
+                                        }
+                                    }
+                                }
+                            }
+                        },
+                        "trainer_advance": {
+                            "total": 1506.1261736038869,
+                            "count": 71154,
+                            "self": 8.044354920290061,
+                            "children": {
+                                "process_trajectory": {
+                                    "total": 323.5915293196158,
+                                    "count": 71154,
+                                    "self": 323.3412643446063,
+                                    "children": {
+                                        "RLTrainer._checkpoint": {
+                                            "total": 0.2502649750094861,
+                                            "count": 2,
+                                            "self": 0.2502649750094861
+                                        }
+                                    }
+                                },
+                                "_update_policy": {
+                                    "total": 1174.490289363981,
+                                    "count": 15,
+                                    "self": 185.54603042497183,
+                                    "children": {
+                                        "TorchPOCAOptimizer.update": {
+                                            "total": 988.9442589390092,
+                                            "count": 1200,
+                                            "self": 988.9442589390092
+                                        }
+                                    }
+                                }
+                            }
+                        }
+                    }
+                },
+                "trainer_threads": {
+                    "total": 8.600036380812526e-07,
+                    "count": 1,
+                    "self": 8.600036380812526e-07
+                },
+                "TrainerController._save_models": {
+                    "total": 0.12173030999838375,
+                    "count": 1,
+                    "self": 0.002498182002454996,
+                    "children": {
+                        "RLTrainer._checkpoint": {
+                            "total": 0.11923212799592875,
+                            "count": 1,
+                            "self": 0.11923212799592875
+                        }
+                    }
+                }
+            }
+        }
+    }
+}

run_logs/training_status.json

@@ -0,0 +1,66 @@
+{
+    "SoccerTwos": {
+        "checkpoints": [
+            {
+                "steps": 4999994,
+                "file_path": "results/SoccerTwos-read-deal/SoccerTwos/SoccerTwos-4999994.onnx",
+                "reward": 0.0,
+                "creation_time": 1710305448.3632262,
+                "auxillary_file_paths": [
+                    "results/SoccerTwos-read-deal/SoccerTwos/SoccerTwos-4999994.pt"
+                ]
+            },
+            {
+                "steps": 5000026,
+                "file_path": "results/SoccerTwos-read-deal/SoccerTwos/SoccerTwos-5000026.onnx",
+                "reward": 0.0,
+                "creation_time": 1710305448.48951,
+                "auxillary_file_paths": [
+                    "results/SoccerTwos-read-deal/SoccerTwos/SoccerTwos-5000026.pt"
+                ]
+            },
+            {
+                "steps": 5499990,
+                "file_path": "results/SoccerTwos-read-deal/SoccerTwos/SoccerTwos-5499990.onnx",
+                "reward": 0.0,
+                "creation_time": 1710334432.864333,
+                "auxillary_file_paths": [
+                    "results/SoccerTwos-read-deal/SoccerTwos/SoccerTwos-5499990.pt"
+                ]
+            },
+            {
+                "steps": 5999976,
+                "file_path": "results/SoccerTwos-read-deal/SoccerTwos/SoccerTwos-5999976.onnx",
+                "reward": 0.0,
+                "creation_time": 1710335864.5550687,
+                "auxillary_file_paths": [
+                    "results/SoccerTwos-read-deal/SoccerTwos/SoccerTwos-5999976.pt"
+                ]
+            },
+            {
+                "steps": 6000104,
+                "file_path": "results/SoccerTwos-read-deal/SoccerTwos/SoccerTwos-6000104.onnx",
+                "reward": 0.0,
+                "creation_time": 1710335864.7100072,
+                "auxillary_file_paths": [
+                    "results/SoccerTwos-read-deal/SoccerTwos/SoccerTwos-6000104.pt"
+                ]
+            }
+        ],
+        "elo": 1561.4600643759927,
+        "final_checkpoint": {
+            "steps": 6000104,
+            "file_path": "results/SoccerTwos-read-deal/SoccerTwos.onnx",
+            "reward": 0.0,
+            "creation_time": 1710335864.7100072,
+            "auxillary_file_paths": [
+                "results/SoccerTwos-read-deal/SoccerTwos/SoccerTwos-6000104.pt"
+            ]
+        }
+    },
+    "metadata": {
+        "stats_format_version": "0.3.0",
+        "mlagents_version": "1.1.0.dev0",
+        "torch_version": "2.2.1+cu121"
+    }
+}