Merge branch 'master' into hh/develop/dodgeball

.github/workflows/pytest.yml

     - 'gym-unity/**'
     - 'test_constraints*.txt'
     - 'test_requirements.txt'
+    - '.github/workflows/pytest.yml'
   push:
     branches: [master]
 
+    env:
+        TEST_ENFORCE_BUFFER_KEY_TYPES: 1
-        include:
-          - python-version: 3.6.x
-            pip_constraints: test_constraints_min_version.txt
-          - python-version: 3.7.x
-            pip_constraints: test_constraints_max_tf1_version.txt
-          - python-version: 3.8.x
-            pip_constraints: test_constraints_max_tf2_version.txt
     steps:
     - uses: actions/checkout@v2
     - name: Set up Python
         # This path is specific to Ubuntu
         path: ~/.cache/pip
         # Look to see if there is a cache hit for the corresponding requirements file
-        key: ${{ runner.os }}-pip-${{ hashFiles('ml-agents/setup.py', 'ml-agents-envs/setup.py', 'gym-unity/setup.py', 'test_requirements.txt', matrix.pip_constraints) }}
+        key: ${{ runner.os }}-pip-${{ hashFiles('ml-agents/setup.py', 'ml-agents-envs/setup.py', 'gym-unity/setup.py', 'test_requirements.txt') }}
         restore-keys: |
           ${{ runner.os }}-pip-
           ${{ runner.os }}-
       run: |
         python -m pip install --upgrade pip
         python -m pip install --upgrade setuptools
-        python -m pip install --progress-bar=off -e ./ml-agents-envs -c ${{ matrix.pip_constraints }}
-        python -m pip install --progress-bar=off -e ./ml-agents -c ${{ matrix.pip_constraints }}
-        python -m pip install --progress-bar=off -r test_requirements.txt -c ${{ matrix.pip_constraints }}
-        python -m pip install --progress-bar=off -e ./gym-unity -c ${{ matrix.pip_constraints }}
+        python -m pip install --progress-bar=off -e ./ml-agents-envs
+        python -m pip install --progress-bar=off -e ./ml-agents
+        python -m pip install --progress-bar=off -r test_requirements.txt
+        python -m pip install --progress-bar=off -e ./gym-unity
+        python -m pip install --progress-bar=off -e ./ml-agents-plugin-examples
     - name: Save python dependencies
       run: |
         pip freeze > pip_versions-${{ matrix.python-version }}.txt

.gitignore

 /summaries
 # Output Artifacts
 /results
+# Output Builds
+/Builds
 
 # Training environments
 /envs

.pre-commit-config.yaml

         args: [--py3-plus, --py36-plus]
         exclude: >
             (?x)^(
-                .*barracuda.py|
                 .*_pb2.py|
                 .*_pb2_grpc.py
             )$
         args: [--assume-in-merge]
     -   id: check-yaml
         # Won't handle the templating in yamato
-        exclude: \.yamato/*
+        exclude: \.yamato/.*
-
--   repo: https://github.com/pre-commit/mirrors-pylint
-    rev: v2.4.4
-    hooks:
-    -   id: pylint
-        exclude: >
-            (?x)^(
-                .*_pb2.py|
-                .*_pb2_grpc.py|
-                .*/tests/.*
-            )$
-        args: [--score=n]
 
 - repo: https://github.com/mattlqx/pre-commit-search-and-replace
   rev: v1.0.3

.yamato/com.unity.ml-agents-pack.yml

 pack:
   name: Pack
   agent:
-    type: Unity::VM::osx
-    image: package-ci/mac:stable
+    type: Unity::VM
+    image: package-ci/ubuntu:stable
     flavor: b1.small
   commands:
     - npm install upm-ci-utils@stable -g --registry https://artifactory.prd.cds.internal.unity3d.com/artifactory/api/npm/upm-npm

.yamato/com.unity.ml-agents-performance.yml

 test_editors:
   - version: 2019.4
-  - version: 2020.1
   - version: 2020.2
 ---
 {% for editor in test_editors %}
   variables:
     UNITY_VERSION: {{ editor.version }}
   commands:
-    - python -m pip install unity-downloader-cli --index-url https://artifactory.prd.it.unity3d.com/artifactory/api/pypi/pypi/simple --upgrade
+    - python3 -m pip install unity-downloader-cli --index-url https://artifactory.prd.it.unity3d.com/artifactory/api/pypi/pypi/simple --upgrade
-    - curl -s https://artifactory.internal.unity3d.com/core-automation/tools/utr-standalone/utr --output utr
+    - curl -s https://artifactory.prd.it.unity3d.com/artifactory/unity-tools-local/utr-standalone/utr --output utr
-    expression: |
-      (pull_request.target eq "master" OR
-      pull_request.target match "release.+") AND
-      NOT pull_request.draft AND
-      (pull_request.changes.any match "com.unity.ml-agents/**" OR
-      pull_request.changes.any match "DevProject/**" OR
-      pull_request.changes.any match "ml-agents/**" OR
-      pull_request.changes.any match "ml-agents-envs/**" OR
-      pull_request.changes.any match ".yamato/com.unity.ml-agents-performance.yml") AND
-      NOT pull_request.changes.all match "**/*.md"
+    recurring:
+      - branch: master
+        frequency: daily
   artifacts:
     logs:
       paths:

.yamato/com.unity.ml-agents-test.yml

     enableCodeCoverage: !!bool false
     # We want some scene tests to run in the DevProject, but packages there only support 2019+
     testProject: Project
+    enableNoDefaultPackages: !!bool false
-  - version: 2020.1
-    enableCodeCoverage: !!bool true
-    testProject: DevProject
+    enableNoDefaultPackages: !!bool true
+    enableNoDefaultPackages: !!bool true
-    enableCodeCoverage: !!bool true
+    # Workaround for MLA-1596 - need to make sure we load the right results.
+    enableCodeCoverage: !!bool false
     testProject: DevProject
 
 test_platforms:
     assembly: Unity.ML-Agents
     minCoveragePct: 72
   - name: com.unity.ml-agents.extensions
-    assembly: Unity.ML-Agents.Extensions
+    assembly: Unity.ML-Agents.Extensions*
     minCoveragePct: 75
 ---
 
 {% else %}
   {% assign coverageOptions = "" %}
 {% endif %}
+
+{% if editor.enableNoDefaultPackages %}
+  {% assign noDefaultPackagesOptions = "--extra-create-project-arg='-upmNoDefaultPackages'" %}
+{% else %}
+  {% assign noDefaultPackagesOptions = "" %}
+{% endif %}
+
 test_{{ package.name }}_{{ platform.name }}_{{ editor.version }}:
   name : {{ package.name }} test {{ editor.version }} on {{ platform.name }}
   agent:
   commands:
     - npm install upm-ci-utils@stable -g --registry https://artifactory.prd.cds.internal.unity3d.com/artifactory/api/npm/upm-npm
-    - upm-ci project test -u {{ editor.version }} --project-path {{ editor.testProject }} --package-filter {{ package.name }} {{ coverageOptions }} --extra-utr-arg "reruncount=2"
+    - upm-ci project test -u {{ editor.version }} --project-path {{ editor.testProject }} --package-filter {{ package.name }} {{ coverageOptions }} {{ noDefaultPackagesOptions }} --extra-utr-arg "reruncount=2"
     {% if editor.enableCodeCoverage %}
     - python3 ml-agents/tests/yamato/check_coverage_percent.py upm-ci~/test-results/ {{ package.minCoveragePct }}
     {% endif %}
     - .yamato/com.unity.ml-agents-pack.yml#pack
   triggers:
     cancel_old_ci: true
-    {% if platform.name == "mac" %}
+    {% if platform.name == "linux" %}
     expression: |
       (pull_request.target eq "master" OR
       pull_request.target match "release.+") AND
     image: {{ platform.image }}
     flavor: {{ platform.flavor}}
   commands:
-    - python -m pip install unity-downloader-cli --index-url https://artifactory.prd.it.unity3d.com/artifactory/api/pypi/pypi/simple --upgrade
+    - python3 -m pip install unity-downloader-cli --index-url https://artifactory.prd.it.unity3d.com/artifactory/api/pypi/pypi/simple --upgrade
-    - upm-ci project test -u {{ editor.version }} --project-path {{ editor.testProject }} --package-filter {{ package.name }} {{ coverageOptions }} --extra-utr-arg "reruncount=2"
+    - upm-ci project test -u {{ editor.version }} --project-path {{ editor.testProject }} --package-filter {{ package.name }} {{ coverageOptions }} --extra-create-project-arg="-upmNoDefaultPackages" --extra-utr-arg "reruncount=2"
     {% if editor.enableCodeCoverage %}
     - python3 ml-agents/tests/yamato/check_coverage_percent.py upm-ci~/test-results/ {{ package.minCoveragePct }}
     {% endif %}

.yamato/compressed-sensor-test.yml

 test_compressed_obs_{{ editor.version }}:
   name: Test Compressed Sensor Observation {{ editor.version }}
   agent:
-    type: Unity::VM::osx
-    image: ml-agents/ml-agents-bokken-mac:0.1.4-492264
-    flavor: b1.small
+    type: Unity::VM
+    image: package-ci/ubuntu:stable
+    flavor: b1.medium
-    - pip install pyyaml --index-url https://artifactory.prd.it.unity3d.com/artifactory/api/pypi/pypi/simple
-    - python -u -m ml-agents.tests.yamato.setup_venv
-    - ./venv/bin/python ml-agents/tests/yamato/scripts/run_compressed_sensor.py --env=artifacts/testPlayer-TestGridCompressed
-    - ./venv/bin/python ml-agents/tests/yamato/scripts/run_compressed_sensor.py --env=artifacts/testPlayer-TestTextureCompressed
+    - |
+      sudo apt-get update && sudo apt-get install -y python3-venv
+      python3 -m venv venv && source venv/bin/activate
+      python -m pip install pyyaml --index-url https://artifactory.prd.it.unity3d.com/artifactory/api/pypi/pypi/simple
+      python -u -m ml-agents.tests.yamato.setup_venv
+      python ml-agents/tests/yamato/scripts/run_compressed_sensor.py --env=artifacts/testPlayer-TestGridCompressed
+      python ml-agents/tests/yamato/scripts/run_compressed_sensor.py --env=artifacts/testPlayer-TestTextureCompressed
-    - .yamato/standalone-build-test.yml#test_mac_standalone_{{ editor.version }}
+    - .yamato/standalone-build-test.yml#test_linux_standalone_{{ editor.version }}
+    {% if editor.extra_test == "sensor" %}
     expression: |
       (pull_request.target eq "master" OR
       pull_request.target match "release.+") AND
       pull_request.changes.any match "Project/**" OR
-      pull_request.changes.any match "ml-agents/**" OR
+      pull_request.changes.any match "ml-agents/tests/yamato/**" OR
+    {% endif %}
 {% endfor %}

.yamato/gym-interface-test.yml

 ---
 {% for editor in test_editors %}
 test_gym_interface_{{ editor.version }}:
-  name: Test Mac Gym Interface {{ editor.version }}
+  name: Test Linux Gym Interface {{ editor.version }}
-    type: Unity::VM::osx
-    image: ml-agents/ml-agents-bokken-mac:0.1.4-492264
-    flavor: b1.small
+    type: Unity::VM
+    image: package-ci/ubuntu:stable
+    flavor: b1.medium
-    - pip install pyyaml --index-url https://artifactory.prd.it.unity3d.com/artifactory/api/pypi/pypi/simple
-    - python -u -m ml-agents.tests.yamato.setup_venv
-    - ./venv/bin/python ml-agents/tests/yamato/scripts/run_gym.py --env=artifacts/testPlayer-Basic
+    - |
+      sudo apt-get update && sudo apt-get install -y python3-venv
+      python3 -m venv venv && source venv/bin/activate
+      python -m pip install wheel --index-url https://artifactory.prd.it.unity3d.com/artifactory/api/pypi/pypi/simple
+      python -m pip install pyyaml --index-url https://artifactory.prd.it.unity3d.com/artifactory/api/pypi/pypi/simple
+      python -u -m ml-agents.tests.yamato.setup_venv
+      python ml-agents/tests/yamato/scripts/run_gym.py --env=artifacts/testPlayer-Basic
-    - .yamato/standalone-build-test.yml#test_mac_standalone_{{ editor.version }}
+    - .yamato/standalone-build-test.yml#test_linux_standalone_{{ editor.version }}
+    {% if editor.extra_test == "gym" %}
     expression: |
       (pull_request.target eq "master" OR
       pull_request.target match "release.+") AND
-      pull_request.changes.any match "ml-agents/**" OR
+      pull_request.changes.any match "ml-agents/tests/yamato/**" OR
+    {% endif %}
 {% endfor %}

.yamato/protobuf-generation-test.yml

-test_mac_protobuf_generation:
+test_linux_protobuf_generation:
-    type: Unity::VM::osx
-    image: ml-agents/ml-agents-bokken-mac:0.1.4-492264
-    flavor: b1.small
+    type: Unity::VM
+    image: package-ci/ubuntu:stable
+    flavor: b1.large
-      brew install nuget
+      sudo apt-get update && sudo apt-get install -y python3-venv nuget
+      python3 -m venv venv && source venv/bin/activate
-      python3 -m venv venv
-      . venv/bin/activate
-      pip install --upgrade pip --index-url https://artifactory.prd.it.unity3d.com/artifactory/api/pypi/pypi/simple
-      pip install grpcio==1.28.1 grpcio-tools==1.13.0 protobuf==3.11.3 six==1.14.0 mypy-protobuf==1.16.0  --progress-bar=off --index-url https://artifactory.prd.it.unity3d.com/artifactory/api/pypi/pypi/simple
-      cd protobuf-definitions
-      chmod +x Grpc.Tools.$GRPC_VERSION/tools/macosx_x64/protoc
-      chmod +x Grpc.Tools.$GRPC_VERSION/tools/macosx_x64/grpc_csharp_plugin
-      COMPILER=Grpc.Tools.$GRPC_VERSION/tools/macosx_x64 ./make.sh
+      python3 -m pip install --upgrade pip --index-url https://artifactory.prd.it.unity3d.com/artifactory/api/pypi/pypi/simple
+      python3 -m pip install grpcio==1.28.1 grpcio-tools==1.13.0 protobuf==3.11.3 six==1.14.0 mypy-protobuf==1.16.0  --progress-bar=off --index-url https://artifactory.prd.it.unity3d.com/artifactory/api/pypi/pypi/simple
+      pushd protobuf-definitions
+      chmod +x Grpc.Tools.$GRPC_VERSION/tools/linux_x64/protoc
+      chmod +x Grpc.Tools.$GRPC_VERSION/tools/linux_x64/grpc_csharp_plugin
+      COMPILER=Grpc.Tools.$GRPC_VERSION/tools/linux_x64 ./make.sh
+      popd
       mkdir -p artifacts
       touch artifacts/proto.patch
       git diff --exit-code -- :/ ":(exclude,top)$CS_PROTO_PATH/*.meta"                          \
       pull_request.target match "release.+") AND
       NOT pull_request.draft AND
       (pull_request.changes.any match "protobuf-definitions/**" OR
+      pull_request.changes.any match "com.unity.ml-agents/Runtime/Grpc/CommunicatorObjects/**" OR
+      pull_request.changes.any match "ml-agents-envs/mlagents_envs/communicator_objects/**" OR
       pull_request.changes.any match ".yamato/protobuf-generation-test.yml") AND
       NOT pull_request.changes.all match "protobuf-definitions/**/*.md"
   artifacts:

.yamato/python-ll-api-test.yml

 {% metadata_file .yamato/test_versions.metafile %}
 ---
 {% for editor in test_editors %}
-test_mac_ll_api_{{ editor.version }}:
-  name: Test Mac LL-API {{ editor.version }}
+test_linux_ll_api_{{ editor.version }}:
+  name: Test Linux LL-API {{ editor.version }}
-    type: Unity::VM::osx
-    image: ml-agents/ml-agents-bokken-mac:0.1.4-492264
-    flavor: b1.small
+    type: Unity::VM
+    image: package-ci/ubuntu:stable
+    flavor: b1.medium
-    - pip install pyyaml --index-url https://artifactory.prd.it.unity3d.com/artifactory/api/pypi/pypi/simple
-    - python -u -m ml-agents.tests.yamato.setup_venv
-    - ./venv/bin/python ml-agents/tests/yamato/scripts/run_llapi.py
-    - ./venv/bin/python ml-agents/tests/yamato/scripts/run_llapi.py --env=artifacts/testPlayer-Basic
-    - ./venv/bin/python ml-agents/tests/yamato/scripts/run_llapi.py --env=artifacts/testPlayer-WallJump
-    - ./venv/bin/python ml-agents/tests/yamato/scripts/run_llapi.py --env=artifacts/testPlayer-Bouncer
+    - |
+      sudo apt-get update && sudo apt-get install -y python3-venv
+      python3 -m venv venv && source venv/bin/activate
+      python -m pip install pyyaml --index-url https://artifactory.prd.it.unity3d.com/artifactory/api/pypi/pypi/simple
+      python -u -m ml-agents.tests.yamato.setup_venv
+      python ml-agents/tests/yamato/scripts/run_llapi.py
+      python ml-agents/tests/yamato/scripts/run_llapi.py --env=artifacts/testPlayer-Basic
+      python ml-agents/tests/yamato/scripts/run_llapi.py --env=artifacts/testPlayer-WallJump
+      python ml-agents/tests/yamato/scripts/run_llapi.py --env=artifacts/testPlayer-Bouncer
-    - .yamato/standalone-build-test.yml#test_mac_standalone_{{ editor.version }}
+    - .yamato/standalone-build-test.yml#test_linux_standalone_{{ editor.version }}
+    {% if editor.extra_test == "llapi" %}
     expression: |
       (pull_request.target eq "master" OR
       pull_request.target match "release.+") AND
-      pull_request.changes.any match "ml-agents/**" OR
+      pull_request.changes.any match "ml-agents/tests/yamato/**" OR
+    {% endif %}
 {% endfor %}

.yamato/standalone-build-test.yml

 {% metadata_file .yamato/test_versions.metafile %}
 ---
 {% for editor in test_editors %}
-test_mac_standalone_{{ editor.version }}:
-  name: Test Mac Standalone {{ editor.version }}
+test_linux_standalone_{{ editor.version }}:
+  name: Test Linux Standalone {{ editor.version }}
-    type: Unity::VM::osx
-    image: ml-agents/ml-agents-bokken-mac:0.1.4-492264
-    flavor: i1.small
+    type: Unity::VM
+    image: package-ci/ubuntu:stable
+    flavor: b1.large
-    - pip install pyyaml --index-url https://artifactory.prd.it.unity3d.com/artifactory/api/pypi/pypi/simple
-    # TODO remove the "--user" command and the path prefix when we can migrate away from the custom bokken image
-    - python -m pip install unity-downloader-cli --index-url https://artifactory.prd.it.unity3d.com/artifactory/api/pypi/pypi/simple --upgrade --user
-    - /Users/bokken/Library/Python/3.7/bin/unity-downloader-cli -u {{ editor.version }} -c editor --wait --fast
-    - python -u -m ml-agents.tests.yamato.standalone_build_tests
-    - python -u -m ml-agents.tests.yamato.standalone_build_tests --scene=Assets/ML-Agents/Examples/Basic/Scenes/Basic.unity
-    - python -u -m ml-agents.tests.yamato.standalone_build_tests --scene=Assets/ML-Agents/Examples/Bouncer/Scenes/Bouncer.unity
-    - python -u -m ml-agents.tests.yamato.standalone_build_tests --scene=Assets/ML-Agents/Examples/WallJump/Scenes/WallJump.unity
-    - python -u -m ml-agents.tests.yamato.standalone_build_tests --scene=Assets/ML-Agents/TestScenes/TestCompressedGrid/TestGridCompressed.unity
-    - python -u -m ml-agents.tests.yamato.standalone_build_tests --scene=Assets/ML-Agents/TestScenes/TestCompressedTexture/TestTextureCompressed.unity
+    - python3 -m pip install pyyaml --index-url https://artifactory.prd.it.unity3d.com/artifactory/api/pypi/pypi/simple
+    - python3 -m pip install unity-downloader-cli --index-url https://artifactory.prd.it.unity3d.com/artifactory/api/pypi/pypi/simple --upgrade
+    - unity-downloader-cli -u {{ editor.version }} -c editor --wait --fast
+    - python3 -u -m ml-agents.tests.yamato.standalone_build_tests --build-target=linux
+    - python3 -u -m ml-agents.tests.yamato.standalone_build_tests --build-target=linux --scene=Assets/ML-Agents/Examples/Basic/Scenes/Basic.unity
+    - python3 -u -m ml-agents.tests.yamato.standalone_build_tests --build-target=linux --scene=Assets/ML-Agents/Examples/Bouncer/Scenes/Bouncer.unity
+    - python3 -u -m ml-agents.tests.yamato.standalone_build_tests --build-target=linux --scene=Assets/ML-Agents/Examples/WallJump/Scenes/WallJump.unity
+    - python3 -u -m ml-agents.tests.yamato.standalone_build_tests --build-target=linux --scene=Assets/ML-Agents/TestScenes/TestCompressedGrid/TestGridCompressed.unity
+    - python3 -u -m ml-agents.tests.yamato.standalone_build_tests --build-target=linux --scene=Assets/ML-Agents/TestScenes/TestCompressedTexture/TestTextureCompressed.unity
   triggers:
     cancel_old_ci: true
     expression: |
     standalonebuild:
       paths:
         - "artifacts/testPlayer*/**"
+        - "artifacts/**/UnityPlayer.so"
 {% endfor %}

.yamato/test_versions.metafile

 # List of editor versions for standalone-build-test and its dependencies.
-# csharp_backcompat_version is used in training-int-tests to determine the
-# older package version to run the backwards compat tests against.
+# We always run training-int-tests for all versions of the editor
+# For each "other" test, we only run it against a single version of the
+# editor to reduce the number of yamato jobs
-    csharp_backcompat_version: 1.0.0
+    extra_test: llapi
-    csharp_backcompat_version: 1.0.0
-  - version: 2020.1
-    csharp_backcompat_version: 1.0.0
+    extra_test: gym
-    # 2020.2 moved the AssetImporters namespace
-    # but we didn't handle this until 1.2.0
-    csharp_backcompat_version: 1.2.0
+    extra_test: sensor

.yamato/training-int-tests.yml

 {% metadata_file .yamato/test_versions.metafile %}
 ---
 {% for editor in test_editors %}
-test_mac_training_int_{{ editor.version }}:
-  name: Test Mac Fast Training {{ editor.version }}
+test_linux_training_int_{{ editor.version }}:
+  name: Test Linux Fast Training {{ editor.version }}
-    type: Unity::VM::osx
-    image: ml-agents/ml-agents-bokken-mac:0.1.4-492264
-    flavor: b1.small
+    type: Unity::VM
+    image: package-ci/ubuntu:stable
+    flavor: b1.medium
-    - pip install pyyaml --index-url https://artifactory.prd.it.unity3d.com/artifactory/api/pypi/pypi/simple
-    # TODO remove the "--user" command and the path prefix when we can migrate away from the custom bokken image
-    - python -m pip install unity-downloader-cli --index-url https://artifactory.prd.it.unity3d.com/artifactory/api/pypi/pypi/simple --upgrade --user
-    - /Users/bokken/Library/Python/3.7/bin/unity-downloader-cli -u {{ editor.version }} -c editor --wait --fast
-    - python -u -m ml-agents.tests.yamato.training_int_tests
-    # Backwards-compatibility tests.
-    # If we make a breaking change to the communication protocol, these will need
-    # to be disabled until the next release.
-    - python -u -m ml-agents.tests.yamato.training_int_tests --python=0.16.0
-    - python -u -m ml-agents.tests.yamato.training_int_tests --csharp={{ editor.csharp_backcompat_version }}
+    - |
+      sudo apt-get update && sudo apt-get install -y python3-venv
+      python3 -m venv venv && source venv/bin/activate
+      python -m pip install pyyaml --index-url https://artifactory.prd.it.unity3d.com/artifactory/api/pypi/pypi/simple
+      python -u -m ml-agents.tests.yamato.training_int_tests
-    - .yamato/standalone-build-test.yml#test_mac_standalone_{{ editor.version }}
+    - .yamato/standalone-build-test.yml#test_linux_standalone_{{ editor.version }}
   triggers:
     cancel_old_ci: true
     expression: |
         - "artifacts/inference.onnx.txt"
     standalonebuild:
       paths:
-        - "artifacts/testplayer*/**"
+        - "artifacts/testPlayer*/**"
         - "artifacts/models/**"
 {% endfor %}

DevProject/Packages/manifest.json

   "dependencies": {
     "com.unity.2d.sprite": "1.0.0",
     "com.unity.2d.tilemap": "1.0.0",
-    "com.unity.ads": "3.4.7",
+    "com.unity.ads": "3.6.1",
-    "com.unity.ide.vscode": "1.2.1",
+    "com.unity.ide.vscode": "1.2.3",
+    "com.unity.inputsystem": "1.1.0-preview.3",
-    "com.unity.multiplayer-hlapi": "1.0.6",
-    "com.unity.package-manager-doctools": "1.1.1-preview.3",
-    "com.unity.package-validation-suite": "0.14.0-preview",
-    "com.unity.purchasing": "2.0.6",
-    "com.unity.test-framework": "1.1.14",
+    "com.unity.multiplayer-hlapi": "1.0.8",
+    "com.unity.package-manager-doctools": "1.7.0-preview",
+    "com.unity.package-validation-suite": "0.19.0-preview",
+    "com.unity.purchasing": "2.2.1",
+    "com.unity.test-framework": "1.1.20",
-    "com.unity.testtools.codecoverage": "0.2.2-preview",
+    "com.unity.testtools.codecoverage": "1.0.0-pre.3",
-    "com.unity.xr.legacyinputhelpers": "2.1.4",
+    "com.unity.xr.legacyinputhelpers": "2.1.7",
     "com.unity.modules.ai": "1.0.0",
     "com.unity.modules.androidjni": "1.0.0",
     "com.unity.modules.animation": "1.0.0",
   "registry": "https://artifactory.prd.cds.internal.unity3d.com/artifactory/api/npm/upm-candidates",
   "testables": [
     "com.unity.ml-agents",
-    "com.unity.ml-agents.extensions"
+    "com.unity.ml-agents.extensions",
+    "com.unity.inputsystem"
   ]
 }

DevProject/ProjectSettings/ProjectSettings.asset

   androidBlitType: 0
   defaultIsNativeResolution: 1
   macRetinaSupport: 1
-  runInBackground: 0
+  runInBackground: 1
   captureSingleScreen: 0
   muteOtherAudioSources: 0
   Prepare IOS For Recording: 0
   xboxOneMonoLoggingLevel: 0
   xboxOneLoggingLevel: 1
   xboxOneDisableEsram: 0
+  xboxOneEnableTypeOptimization: 0
   xboxOnePresentImmediateThreshold: 0
   switchQueueCommandMemory: 1048576
   switchQueueControlMemory: 16384
   switchNVNOtherPoolsGranularity: 16777216
+  switchNVNMaxPublicTextureIDCount: 0
+  switchNVNMaxPublicSamplerIDCount: 0
+  stadiaPresentMode: 0
+  stadiaTargetFramerate: 0
+  vulkanEnableLateAcquireNextImage: 0
   m_SupportedAspectRatios:
     4:3: 1
     5:4: 1
   useHDRDisplay: 0
   D3DHDRBitDepth: 0
   m_ColorGamuts: 00000000
-  targetPixelDensity: 0
+  targetPixelDensity: 30
   resolutionScalingMode: 0
   androidSupportedAspectRatio: 1
   androidMaxAspectRatio: 2.1
   StripUnusedMeshComponents: 0
   VertexChannelCompressionMask: 4054
   iPhoneSdkVersion: 988
-  iOSTargetOSVersionString: 
+  iOSTargetOSVersionString: 10.0
-  tvOSTargetOSVersionString: 
+  tvOSTargetOSVersionString: 10.0
   uIPrerenderedIcon: 0
   uIRequiresPersistentWiFi: 0
   uIRequiresFullScreen: 1
-  iPhoneSplashScreen: {fileID: 0}
-  iPhoneHighResSplashScreen: {fileID: 0}
-  iPhoneTallHighResSplashScreen: {fileID: 0}
-  iPhone47inSplashScreen: {fileID: 0}
-  iPhone55inPortraitSplashScreen: {fileID: 0}
-  iPhone55inLandscapeSplashScreen: {fileID: 0}
-  iPhone58inPortraitSplashScreen: {fileID: 0}
-  iPhone58inLandscapeSplashScreen: {fileID: 0}
-  iPadPortraitSplashScreen: {fileID: 0}
-  iPadHighResPortraitSplashScreen: {fileID: 0}
-  iPadLandscapeSplashScreen: {fileID: 0}
-  iPadHighResLandscapeSplashScreen: {fileID: 0}
-  iPhone65inPortraitSplashScreen: {fileID: 0}
-  iPhone65inLandscapeSplashScreen: {fileID: 0}
-  iPhone61inPortraitSplashScreen: {fileID: 0}
-  iPhone61inLandscapeSplashScreen: {fileID: 0}
   appleTVSplashScreen: {fileID: 0}
   appleTVSplashScreen2x: {fileID: 0}
   tvOSSmallIconLayers: []
   metalEditorSupport: 1
   metalAPIValidation: 1
   iOSRenderExtraFrameOnPause: 0
+  iosCopyPluginsCodeInsteadOfSymlink: 0
   appleDeveloperTeamID: 
   iOSManualSigningProvisioningProfileID: 
   tvOSManualSigningProvisioningProfileID: 
   ps4ShareFilePath: 
   ps4ShareOverlayImagePath: 
   ps4PrivacyGuardImagePath: 
+  ps4ExtraSceSysFile: 
   ps4NPtitleDatPath: 
   ps4RemotePlayKeyAssignment: -1
   ps4RemotePlayKeyMappingDir: 
   ps4UseResolutionFallback: 0
   ps4ReprojectionSupport: 0
   ps4UseAudio3dBackend: 0
+  ps4UseLowGarlicFragmentationMode: 1
   ps4SocialScreenEnabled: 0
   ps4ScriptOptimizationLevel: 2
   ps4Audio3dVirtualSpeakerCount: 14
   ps4disableAutoHideSplash: 0
   ps4videoRecordingFeaturesUsed: 0
   ps4contentSearchFeaturesUsed: 0
+  ps4CompatibilityPS5: 0
+  ps4GPU800MHz: 1
   ps4attribEyeToEyeDistanceSettingVR: 0
   ps4IncludedModules: []
   ps4attribVROutputEnabled: 0
   additionalIl2CppArgs: 
   scriptingRuntimeVersion: 1
   gcIncremental: 0
+  assemblyVersionValidation: 1
   gcWBarrierValidation: 0
   apiCompatibilityLevelPerPlatform: {}
   m_RenderingPath: 1
   XboxOneCapability: []
   XboxOneGameRating: {}
   XboxOneIsContentPackage: 0
+  XboxOneEnhancedXboxCompatibilityMode: 0
   XboxOneEnableGPUVariability: 1
   XboxOneSockets: {}
   XboxOneSplashScreen: {fileID: 0}
   XboxOneOverrideIdentityName: 
+  XboxOneOverrideIdentityPublisher: 
   vrEditorSettings:
     daydream:
       daydreamIconForeground: {fileID: 0}
   projectName: 
   organizationId: 
   cloudEnabled: 0
-  enableNativePlatformBackendsForNewInputSystem: 0
-  disableOldInputManagerSupport: 0
+  enableNativePlatformBackendsForNewInputSystem: 1
+  disableOldInputManagerSupport: 1
   legacyClampBlendShapeWeights: 0

DevProject/ProjectSettings/ProjectVersion.txt

-m_EditorVersion: 2019.4.7f1
-m_EditorVersionWithRevision: 2019.4.7f1 (e992b1a16e65)
+m_EditorVersion: 2019.4.19f1
+m_EditorVersionWithRevision: 2019.4.19f1 (ca5b14067cec)

Dockerfile

-FROM nvidia/cuda:10.0-cudnn7-devel-ubuntu18.04
+# From https://gitlab.com/nvidia/container-images/cuda/blob/master/doc/supported-tags.md
+FROM nvidia/cuda:10.2-cudnn7-devel-ubuntu18.04
 
 RUN yes | unminimize
 
   apt-get install -y --no-install-recommends wget curl tmux vim git gdebi-core \
   build-essential python3-pip unzip google-cloud-sdk htop mesa-utils xorg-dev xorg \
-  libglvnd-dev libgl1-mesa-dev libegl1-mesa-dev libgles2-mesa-dev && \
+  libglvnd-dev libgl1-mesa-dev libegl1-mesa-dev libgles2-mesa-dev xvfb && \
-  wget http://security.ubuntu.com/ubuntu/pool/universe/x/xorg-server/xvfb_1.18.4-0ubuntu0.10_amd64.deb && \
-  yes | gdebi libxfont1_1.5.1-1ubuntu0.16.04.4_amd64.deb && \
-  yes | gdebi xvfb_1.18.4-0ubuntu0.10_amd64.deb
+  yes | gdebi libxfont1_1.5.1-1ubuntu0.16.04.4_amd64.deb
 RUN python3 -m pip install --upgrade pip
 RUN pip install setuptools==41.0.0
 

Project/Assets/ML-Agents/Editor/Tests/StandaloneBuildTest.cs

     {
         const string k_OutputCommandLineFlag = "--mlagents-build-output-path";
         const string k_SceneCommandLineFlag = "--mlagents-build-scene-path";
+        private const string k_BuildTargetFlag = "--mlagents-build-target";
 
         public static void BuildStandalonePlayerOSX()
         {
+            var buildTarget = BuildTarget.StandaloneOSX;
 
             var args = Environment.GetCommandLineArgs();
             for (var i = 0; i < args.Length - 1; i++)
                 {
                     scenePath = args[i + 1];
                 }
+                else if (args[i] == k_BuildTargetFlag)
+                {
+                    buildTarget = (BuildTarget)Enum.Parse(typeof(BuildTarget), args[i + 1], ignoreCase: true);
+                }
             }
 
             string[] scenes = { scenePath };
-                BuildTarget.StandaloneOSX,
-                BuildOptions.None
+                buildTarget,
+                BuildOptions.Development
             );
             var isOk = buildResult.summary.result == BuildResult.Succeeded;
             var error = "";

Project/Assets/ML-Agents/Examples/3DBall/Prefabs/3DBall.prefab

   m_BrainParameters:
     VectorObservationSize: 8
     NumStackedVectorObservations: 1
+    m_ActionSpec:
+      m_NumContinuousActions: 2
+      BranchSizes: 
+    hasUpgradedBrainParametersWithActionSpec: 1
-  m_InferenceDevice: 0
+  m_InferenceDevice: 2
   m_BehaviorType: 0
   m_BehaviorName: 3DBall
   TeamId: 0

Project/Assets/ML-Agents/Examples/3DBall/Prefabs/3DBallHardNew.prefab

   m_Name: 
   m_EditorClassIdentifier: 
   m_BrainParameters:
-    vectorObservationSize: 5
-    numStackedVectorObservations: 9
-    vectorActionSize: 02000000
-    vectorActionDescriptions: []
-    vectorActionSpaceType: 1
-  m_Model: {fileID: 11400000, guid: 27d49984757ed46b181090a532ef48e5, type: 3}
-  m_InferenceDevice: 0
+    VectorObservationSize: 0
+    NumStackedVectorObservations: 1
+    m_ActionSpec:
+      m_NumContinuousActions: 2
+      BranchSizes: 
+    VectorActionSize: 02000000
+    VectorActionDescriptions: []
+    VectorActionSpaceType: 1
+    hasUpgradedBrainParametersWithActionSpec: 1
+  m_Model: {fileID: 11400000, guid: d179c44c147aa4ffbbb725f009eca3b8, type: 3}
+  m_InferenceDevice: 2
+  m_UseChildActuators: 1
+  m_ObservableAttributeHandling: 1
 --- !u!114 &114466000339026140
 MonoBehaviour:
   m_ObjectHideFlags: 0
   agentParameters:
     maxStep: 0
   hasUpgradedFromAgentParameters: 1
-  maxStep: 5000
+  MaxStep: 5000
   ball: {fileID: 1142513601053358}
 --- !u!114 &8193279139064749781
 MonoBehaviour:
   m_EditorClassIdentifier: 
   DecisionPeriod: 5
   TakeActionsBetweenDecisions: 1
-  offsetStep: 0
 --- !u!114 &7923264721978289873
 MonoBehaviour:
   m_ObjectHideFlags: 0
   m_Script: {fileID: 11500000, guid: 3a6da8f78a394c6ab027688eab81e04d, type: 3}
   m_Name: 
   m_EditorClassIdentifier: 
+  debugCommandLineOverride: 
 --- !u!1 &1978072206102878
 GameObject:
   m_ObjectHideFlags: 0

Project/Assets/ML-Agents/Examples/3DBall/Prefabs/Visual3DBall.prefab

   m_BrainParameters:
     VectorObservationSize: 0
     NumStackedVectorObservations: 1
+    m_ActionSpec:
+      m_NumContinuousActions: 2
+      BranchSizes: 
+    hasUpgradedBrainParametersWithActionSpec: 1
-  m_InferenceDevice: 0
+  m_InferenceDevice: 2
   m_BehaviorType: 0
   m_BehaviorName: Visual3DBall
   TeamId: 0

Project/Assets/ML-Agents/Examples/3DBall/Scripts/Ball3DAgent.cs

-using System;
 using UnityEngine;
 using Unity.MLAgents;
 using Unity.MLAgents.Actuators;

Project/Assets/ML-Agents/Examples/3DBall/Scripts/Ball3DHardAgent.cs

 using Unity.MLAgents;
 using Unity.MLAgents.Actuators;
 using Unity.MLAgents.Sensors;
+using Unity.MLAgents.Sensors.Reflection;
 
 public class Ball3DHardAgent : Agent
 {
         SetResetParameters();
     }
 
-    public override void CollectObservations(VectorSensor sensor)
+    [Observable(numStackedObservations: 9)]
+    Vector2 Rotation
-        sensor.AddObservation(gameObject.transform.rotation.z);
-        sensor.AddObservation(gameObject.transform.rotation.x);
-        sensor.AddObservation((ball.transform.position - gameObject.transform.position));
+        get
+        {
+            return new Vector2(gameObject.transform.rotation.z, gameObject.transform.rotation.x);
+        }
+    }
+
+    [Observable(numStackedObservations: 9)]
+    Vector3 PositionDelta
+    {
+        get
+        {
+            return ball.transform.position - gameObject.transform.position;
+        }
     }
 
     public override void OnActionReceived(ActionBuffers actionBuffers)

Project/Assets/ML-Agents/Examples/Basic/Prefabs/Basic.prefab

   m_BrainParameters:
     VectorObservationSize: 0
     NumStackedVectorObservations: 1
+    m_ActionSpec:
+      m_NumContinuousActions: 0
+      BranchSizes: 
+    hasUpgradedBrainParametersWithActionSpec: 1
-  m_InferenceDevice: 0
+  m_InferenceDevice: 2
   m_BehaviorType: 0
   m_BehaviorName: Basic
   TeamId: 0

Project/Assets/ML-Agents/Examples/Basic/Scripts/BasicActuatorComponent.cs

 using System;
 using Unity.MLAgents.Actuators;
+using UnityEngine;
 
 namespace Unity.MLAgentsExamples
 {
         /// Creates a BasicActuator.
         /// </summary>
         /// <returns></returns>
+#pragma warning disable 672
+#pragma warning restore 672
         {
             return new BasicActuator(basicController);
         }
     /// <summary>
     /// Simple actuator that converts the action into a {-1, 0, 1} direction
     /// </summary>
-    public class BasicActuator : IActuator
+    public class BasicActuator : IActuator, IHeuristicProvider
     {
         public BasicController basicController;
         ActionSpec m_ActionSpec;
             }
 
             basicController.MoveDirection(direction);
+        }
+
+        public void Heuristic(in ActionBuffers actionBuffersOut)
+        {
+            var direction = Input.GetAxis("Horizontal");
+            var discreteActions = actionBuffersOut.DiscreteActions;
+            if (Mathf.Approximately(direction, 0.0f))
+            {
+                discreteActions[0] = 0;
+                return;
+            }
+            var sign = Math.Sign(direction);
+            discreteActions[0] = sign < 0 ? 1 : 2;
         }
 
         public void WriteDiscreteActionMask(IDiscreteActionMask actionMask)

Project/Assets/ML-Agents/Examples/Basic/Scripts/BasicController.cs

 using UnityEngine;
 using UnityEngine.SceneManagement;
 using Unity.MLAgents;
-using Unity.MLAgents.Actuators;
 using UnityEngine.Serialization;
 
 /// <summary>

Project/Assets/ML-Agents/Examples/Bouncer/Prefabs/Environment.prefab

   m_Name: 
   m_EditorClassIdentifier: 
   m_BrainParameters:
-    vectorObservationSize: 6
-    numStackedVectorObservations: 3
-    vectorActionSize: 03000000
-    vectorActionDescriptions: []
-    vectorActionSpaceType: 1
+    VectorObservationSize: 6
+    NumStackedVectorObservations: 3
+    m_ActionSpec:
+      m_NumContinuousActions: 3
+      BranchSizes: 
+    VectorActionSize: 03000000
+    VectorActionDescriptions: []
+    VectorActionSpaceType: 1
+    hasUpgradedBrainParametersWithActionSpec: 1
-  m_InferenceDevice: 0
+  m_InferenceDevice: 2
+  m_UseChildActuators: 1
+  m_ObservableAttributeHandling: 0
 --- !u!114 &114878620968301562
 MonoBehaviour:
   m_ObjectHideFlags: 0
   agentParameters:
     maxStep: 0
   hasUpgradedFromAgentParameters: 1
-  maxStep: 0
+  MaxStep: 0
   target: {fileID: 1160631129428284}
   bodyObject: {fileID: 1680588139522898}
   strength: 500
   m_Script: {fileID: 11500000, guid: 3a6da8f78a394c6ab027688eab81e04d, type: 3}
   m_Name: 
   m_EditorClassIdentifier: 
+  debugCommandLineOverride: 
 --- !u!1 &1680588139522898
 GameObject:
   m_ObjectHideFlags: 0

Project/Assets/ML-Agents/Examples/Crawler/Prefabs/CrawlerBase.prefab

   m_BrainParameters:
     VectorObservationSize: 32
     NumStackedVectorObservations: 1
+    m_ActionSpec:
+      m_NumContinuousActions: 20
+      BranchSizes: 
+    hasUpgradedBrainParametersWithActionSpec: 1
-  m_InferenceDevice: 0
+  m_InferenceDevice: 2
   m_BehaviorType: 0
   m_BehaviorName: 
   TeamId: 0

Project/Assets/ML-Agents/Examples/Crawler/Scripts/CrawlerAgent.cs

-using System;
 using UnityEngine;
 using Unity.MLAgents;
 using Unity.Barracuda;
         Vector3 avgVel = Vector3.zero;
 
         //ALL RBS
-        int numOfRB = 0;
+        int numOfRb = 0;
-            numOfRB++;
+            numOfRb++;
-        avgVel = velSum / numOfRB;
+        avgVel = velSum / numOfRb;
         return avgVel;
     }
 

Project/Assets/ML-Agents/Examples/FoodCollector/Prefabs/FoodCollectorArea.prefab

   m_BrainParameters:
     VectorObservationSize: 4
     NumStackedVectorObservations: 1
-    VectorActionSize: 03000000030000000300000002000000
+    m_ActionSpec:
+      m_NumContinuousActions: 3
+      BranchSizes: 02000000
+    VectorActionSize: 
-    VectorActionSpaceType: 0
-  m_Model: {fileID: 11400000, guid: 36ab3e93020504f48858d0856f939685, type: 3}
-  m_InferenceDevice: 0
+    VectorActionSpaceType: 1
+    hasUpgradedBrainParametersWithActionSpec: 1
+  m_Model: {fileID: 11400000, guid: 3210b528a2bc44a86bd6bd1d571070f8, type: 3}
+  m_InferenceDevice: 2
+  m_UseChildActuators: 1
   m_ObservableAttributeHandling: 0
 --- !u!114 &114176228333253036
 MonoBehaviour:
   myLaser: {fileID: 1081721624670010}
   contribute: 1
   useVectorObs: 1
+  useVectorFrozenFlag: 0
 --- !u!114 &114725457980523372
 MonoBehaviour:
   m_ObjectHideFlags: 0
   m_BrainParameters:
     VectorObservationSize: 4
     NumStackedVectorObservations: 1
-    VectorActionSize: 03000000030000000300000002000000
+    m_ActionSpec:
+      m_NumContinuousActions: 3
+      BranchSizes: 02000000
+    VectorActionSize: 
-    VectorActionSpaceType: 0
-  m_Model: {fileID: 11400000, guid: 36ab3e93020504f48858d0856f939685, type: 3}
-  m_InferenceDevice: 0
+    VectorActionSpaceType: 1
+    hasUpgradedBrainParametersWithActionSpec: 1
+  m_Model: {fileID: 11400000, guid: 3210b528a2bc44a86bd6bd1d571070f8, type: 3}
+  m_InferenceDevice: 2
+  m_UseChildActuators: 1
   m_ObservableAttributeHandling: 0
 --- !u!114 &114711827726849508
 MonoBehaviour:
   myLaser: {fileID: 1941433838307300}
   contribute: 0
   useVectorObs: 1
+  useVectorFrozenFlag: 0
 --- !u!114 &114443152683847924
 MonoBehaviour:
   m_ObjectHideFlags: 0
   m_BrainParameters:
     VectorObservationSize: 4
     NumStackedVectorObservations: 1
-    VectorActionSize: 03000000030000000300000002000000
+    m_ActionSpec:
+      m_NumContinuousActions: 3
+      BranchSizes: 02000000
+    VectorActionSize: 
-    VectorActionSpaceType: 0
-  m_Model: {fileID: 11400000, guid: 36ab3e93020504f48858d0856f939685, type: 3}
-  m_InferenceDevice: 0
+    VectorActionSpaceType: 1
+    hasUpgradedBrainParametersWithActionSpec: 1
+  m_Model: {fileID: 11400000, guid: 3210b528a2bc44a86bd6bd1d571070f8, type: 3}
+  m_InferenceDevice: 2
+  m_UseChildActuators: 1
   m_ObservableAttributeHandling: 0
 --- !u!114 &114542632553128056
 MonoBehaviour:
   myLaser: {fileID: 1421240237750412}
   contribute: 0
   useVectorObs: 1
+  useVectorFrozenFlag: 0
 --- !u!114 &114986980423924774
 MonoBehaviour:
   m_ObjectHideFlags: 0
   m_BrainParameters:
     VectorObservationSize: 4
     NumStackedVectorObservations: 1
-    VectorActionSize: 03000000030000000300000002000000
+    m_ActionSpec:
+      m_NumContinuousActions: 3
+      BranchSizes: 02000000
+    VectorActionSize: 
-    VectorActionSpaceType: 0
-  m_Model: {fileID: 11400000, guid: 36ab3e93020504f48858d0856f939685, type: 3}
-  m_InferenceDevice: 0
+    VectorActionSpaceType: 1
+    hasUpgradedBrainParametersWithActionSpec: 1
+  m_Model: {fileID: 11400000, guid: 3210b528a2bc44a86bd6bd1d571070f8, type: 3}
+  m_InferenceDevice: 2
+  m_UseChildActuators: 1
   m_ObservableAttributeHandling: 0
 --- !u!114 &114189751434580810
 MonoBehaviour:
   myLaser: {fileID: 1617924810425504}
   contribute: 0
   useVectorObs: 1
+  useVectorFrozenFlag: 0
 --- !u!114 &114644889237473510
 MonoBehaviour:
   m_ObjectHideFlags: 0
   m_BrainParameters:
     VectorObservationSize: 4
     NumStackedVectorObservations: 1
-    VectorActionSize: 03000000030000000300000002000000
+    m_ActionSpec:
+      m_NumContinuousActions: 3
+      BranchSizes: 02000000
+    VectorActionSize: 
-    VectorActionSpaceType: 0
-  m_Model: {fileID: 11400000, guid: 36ab3e93020504f48858d0856f939685, type: 3}
-  m_InferenceDevice: 0
+    VectorActionSpaceType: 1
+    hasUpgradedBrainParametersWithActionSpec: 1
+  m_Model: {fileID: 11400000, guid: 3210b528a2bc44a86bd6bd1d571070f8, type: 3}
+  m_InferenceDevice: 2
+  m_UseChildActuators: 1
   m_ObservableAttributeHandling: 0
 --- !u!114 &114235147148547996
 MonoBehaviour:
   myLaser: {fileID: 1045923826166930}
   contribute: 0
   useVectorObs: 1
+  useVectorFrozenFlag: 0
 --- !u!114 &114276061479012222
 MonoBehaviour:
   m_ObjectHideFlags: 0
   m_PrefabInstance: {fileID: 0}
   m_PrefabAsset: {fileID: 0}
   m_GameObject: {fileID: 1819751139121548}
-  m_LocalRotation: {x: -0, y: -0, z: -0, w: 1}
+  m_LocalRotation: {x: 0, y: 0, z: 0, w: 1}
   m_LocalPosition: {x: 0, y: 12.3, z: 0}
   m_LocalScale: {x: 1, y: 1, z: 1}
   m_Children:

Project/Assets/ML-Agents/Examples/FoodCollector/Prefabs/GridFoodCollectorArea.prefab

   m_BrainParameters:
     VectorObservationSize: 0
     NumStackedVectorObservations: 1
-    VectorActionSize: 03000000030000000300000002000000
+    m_ActionSpec:
+      m_NumContinuousActions: 3
+      BranchSizes: 02000000
+    VectorActionSize: 
-  m_Model: {fileID: 11400000, guid: 699f852e79b5ba642871514fb1fb9843, type: 3}
-  m_InferenceDevice: 0
+    hasUpgradedBrainParametersWithActionSpec: 1
+  m_Model: {fileID: 11400000, guid: 75910f45f20be49b18e2b95879a217b2, type: 3}
+  m_InferenceDevice: 2
   m_BehaviorType: 0
   m_BehaviorName: GridFoodCollector
   TeamId: 0
   myLaser: {fileID: 1081721624670010}
   contribute: 0
   useVectorObs: 0
+  useVectorFrozenFlag: 0
 --- !u!114 &8297075921230369060
 MonoBehaviour:
   m_ObjectHideFlags: 0
   - {r: 0, g: 0, b: 0, a: 0}
   GizmoYOffset: 0
   ShowGizmos: 0
+  CompressionType: 1
 --- !u!1 &1482701732800114
 GameObject:
   m_ObjectHideFlags: 0
   m_BrainParameters:
     VectorObservationSize: 0
     NumStackedVectorObservations: 1
-    VectorActionSize: 03000000030000000300000002000000
+    m_ActionSpec:
+      m_NumContinuousActions: 3
+      BranchSizes: 02000000
+    VectorActionSize: 
-  m_Model: {fileID: 11400000, guid: 699f852e79b5ba642871514fb1fb9843, type: 3}
-  m_InferenceDevice: 0
+    hasUpgradedBrainParametersWithActionSpec: 1
+  m_Model: {fileID: 11400000, guid: 75910f45f20be49b18e2b95879a217b2, type: 3}
+  m_InferenceDevice: 2
   m_BehaviorType: 0
   m_BehaviorName: GridFoodCollector
   TeamId: 0
   myLaser: {fileID: 1941433838307300}
   contribute: 0
   useVectorObs: 0
+  useVectorFrozenFlag: 0
 --- !u!114 &259154752087955944
 MonoBehaviour:
   m_ObjectHideFlags: 0
   - {r: 0, g: 0, b: 0, a: 0}
   GizmoYOffset: 0
   ShowGizmos: 0
+  CompressionType: 1
 --- !u!1 &1528397385587768
 GameObject:
   m_ObjectHideFlags: 0
   m_BrainParameters:
     VectorObservationSize: 0
     NumStackedVectorObservations: 1
-    VectorActionSize: 03000000030000000300000002000000
+    m_ActionSpec:
+      m_NumContinuousActions: 3
+      BranchSizes: 02000000
+    VectorActionSize: 
-  m_Model: {fileID: 11400000, guid: 699f852e79b5ba642871514fb1fb9843, type: 3}
-  m_InferenceDevice: 0
+    hasUpgradedBrainParametersWithActionSpec: 1
+  m_Model: {fileID: 11400000, guid: 75910f45f20be49b18e2b95879a217b2, type: 3}
+  m_InferenceDevice: 2
   m_BehaviorType: 0
   m_BehaviorName: GridFoodCollector
   TeamId: 0
   myLaser: {fileID: 1421240237750412}
   contribute: 0
   useVectorObs: 0
+  useVectorFrozenFlag: 0
 --- !u!114 &5519119940433428255
 MonoBehaviour:
   m_ObjectHideFlags: 0
   - {r: 0, g: 0, b: 0, a: 0}
   GizmoYOffset: 0
   ShowGizmos: 0
+  CompressionType: 1
 --- !u!1 &1617924810425504
 GameObject:
   m_ObjectHideFlags: 0
   m_BrainParameters:
     VectorObservationSize: 0
     NumStackedVectorObservations: 1
-    VectorActionSize: 03000000030000000300000002000000
+    m_ActionSpec:
+      m_NumContinuousActions: 3
+      BranchSizes: 02000000
+    VectorActionSize: 
-  m_Model: {fileID: 11400000, guid: 699f852e79b5ba642871514fb1fb9843, type: 3}
-  m_InferenceDevice: 0
+    hasUpgradedBrainParametersWithActionSpec: 1
+  m_Model: {fileID: 11400000, guid: 75910f45f20be49b18e2b95879a217b2, type: 3}
+  m_InferenceDevice: 2
   m_BehaviorType: 0
   m_BehaviorName: GridFoodCollector
   TeamId: 0
   myLaser: {fileID: 1617924810425504}
   contribute: 0
   useVectorObs: 0
+  useVectorFrozenFlag: 0
 --- !u!114 &5884750436653390196
 MonoBehaviour:
   m_ObjectHideFlags: 0
   - {r: 0, g: 0, b: 0, a: 0}
   GizmoYOffset: 0
   ShowGizmos: 0
+  CompressionType: 1
 --- !u!1 &1688105343773098
 GameObject:
   m_ObjectHideFlags: 0
   m_BrainParameters:
     VectorObservationSize: 0
     NumStackedVectorObservations: 1
-    VectorActionSize: 03000000030000000300000002000000
+    m_ActionSpec:
+      m_NumContinuousActions: 3
+      BranchSizes: 02000000
+    VectorActionSize: 
-  m_Model: {fileID: 11400000, guid: 699f852e79b5ba642871514fb1fb9843, type: 3}
-  m_InferenceDevice: 0
+    hasUpgradedBrainParametersWithActionSpec: 1
+  m_Model: {fileID: 11400000, guid: 75910f45f20be49b18e2b95879a217b2, type: 3}
+  m_InferenceDevice: 2
   m_BehaviorType: 0
   m_BehaviorName: GridFoodCollector
   TeamId: 0
   myLaser: {fileID: 1045923826166930}
   contribute: 0
   useVectorObs: 0
+  useVectorFrozenFlag: 0
 --- !u!114 &4768752321433982785
 MonoBehaviour:
   m_ObjectHideFlags: 0
   - {r: 0, g: 0, b: 0, a: 0}
   GizmoYOffset: 0
   ShowGizmos: 0
+  CompressionType: 1
 --- !u!1 &1729825611722018
 GameObject:
   m_ObjectHideFlags: 0
   m_PrefabInstance: {fileID: 0}
   m_PrefabAsset: {fileID: 0}
   m_GameObject: {fileID: 1819751139121548}
-  m_LocalRotation: {x: -0, y: -0, z: -0, w: 1}
+  m_LocalRotation: {x: 0, y: 0, z: 0, w: 1}
   m_LocalPosition: {x: 0, y: 12.3, z: 0}
   m_LocalScale: {x: 1, y: 1, z: 1}
   m_Children:

Project/Assets/ML-Agents/Examples/FoodCollector/Prefabs/VisualFoodCollectorArea.prefab

   m_BrainParameters:
     VectorObservationSize: 1
     NumStackedVectorObservations: 1
-    VectorActionSize: 03000000030000000300000002000000
+    m_ActionSpec:
+      m_NumContinuousActions: 3
+      BranchSizes: 02000000
+    VectorActionSize: 
-  m_Model: {fileID: 11400000, guid: c3b1eb0bcf06b4c0488599c7ab806de7, type: 3}
-  m_InferenceDevice: 0
+    hasUpgradedBrainParametersWithActionSpec: 1
+  m_Model: {fileID: 11400000, guid: ec4b31b5d66ca4e51ae3ac41945facb2, type: 3}
+  m_InferenceDevice: 2
   m_BehaviorType: 0
   m_BehaviorName: VisualFoodCollector
   TeamId: 0
   m_BrainParameters:
     VectorObservationSize: 1
     NumStackedVectorObservations: 1
-    VectorActionSize: 03000000030000000300000002000000
+    m_ActionSpec:
+      m_NumContinuousActions: 3
+      BranchSizes: 02000000
+    VectorActionSize: 
-  m_Model: {fileID: 11400000, guid: c3b1eb0bcf06b4c0488599c7ab806de7, type: 3}
-  m_InferenceDevice: 0
+    hasUpgradedBrainParametersWithActionSpec: 1
+  m_Model: {fileID: 11400000, guid: ec4b31b5d66ca4e51ae3ac41945facb2, type: 3}
+  m_InferenceDevice: 2
   m_BehaviorType: 0
   m_BehaviorName: VisualFoodCollector
   TeamId: 0
   m_BrainParameters:
     VectorObservationSize: 1
     NumStackedVectorObservations: 1
-    VectorActionSize: 03000000030000000300000002000000
+    m_ActionSpec:
+      m_NumContinuousActions: 3
+      BranchSizes: 02000000
+    VectorActionSize: 
-  m_Model: {fileID: 11400000, guid: c3b1eb0bcf06b4c0488599c7ab806de7, type: 3}
-  m_InferenceDevice: 0
+    hasUpgradedBrainParametersWithActionSpec: 1
+  m_Model: {fileID: 11400000, guid: ec4b31b5d66ca4e51ae3ac41945facb2, type: 3}
+  m_InferenceDevice: 2
   m_BehaviorType: 0
   m_BehaviorName: VisualFoodCollector
   TeamId: 0
   m_BrainParameters:
     VectorObservationSize: 1
     NumStackedVectorObservations: 1
-    VectorActionSize: 03000000030000000300000002000000
+    m_ActionSpec:
+      m_NumContinuousActions: 3
+      BranchSizes: 02000000
+    VectorActionSize: 
-  m_Model: {fileID: 11400000, guid: c3b1eb0bcf06b4c0488599c7ab806de7, type: 3}
-  m_InferenceDevice: 0
+    hasUpgradedBrainParametersWithActionSpec: 1
+  m_Model: {fileID: 11400000, guid: ec4b31b5d66ca4e51ae3ac41945facb2, type: 3}
+  m_InferenceDevice: 2
   m_BehaviorType: 0
   m_BehaviorName: VisualFoodCollector
   TeamId: 0

Project/Assets/ML-Agents/Examples/FoodCollector/Scripts/FoodCollectorAgent.cs

-using System;
 using UnityEngine;
 using Unity.MLAgents;
 using Unity.MLAgents.Actuators;
         return new Color32(r, g, b, 255);
     }
 
-    public void MoveAgent(ActionSegment<int> act)
+    public void MoveAgent(ActionBuffers actionBuffers)
     {
         m_Shoot = false;
 
         var dirToGo = Vector3.zero;
         var rotateDir = Vector3.zero;
 
+        var continuousActions = actionBuffers.ContinuousActions;
+        var discreteActions = actionBuffers.DiscreteActions;
+
-            var shootCommand = false;
-            var forwardAxis = (int)act[0];
-            var rightAxis = (int)act[1];
-            var rotateAxis = (int)act[2];
-            var shootAxis = (int)act[3];
-
-            switch (forwardAxis)
-            {
-                case 1:
-                    dirToGo = transform.forward;
-                    break;
-                case 2:
-                    dirToGo = -transform.forward;
-                    break;
-            }
+            var forward = Mathf.Clamp(continuousActions[0], -1f, 1f);
+            var right = Mathf.Clamp(continuousActions[1], -1f, 1f);
+            var rotate = Mathf.Clamp(continuousActions[2], -1f, 1f);
-            switch (rightAxis)
-            {
-                case 1:
-                    dirToGo = transform.right;
-                    break;
-                case 2:
-                    dirToGo = -transform.right;
-                    break;
-            }
+            dirToGo = transform.forward * forward;
+            dirToGo += transform.right * right;
+            rotateDir = -transform.up * rotate;
-            switch (rotateAxis)
-            {
-                case 1:
-                    rotateDir = -transform.up;
-                    break;
-                case 2:
-                    rotateDir = transform.up;
-                    break;
-            }
-            switch (shootAxis)
-            {
-                case 1:
-                    shootCommand = true;
-                    break;
-            }
+            var shootCommand = (int)discreteActions[0] > 0;
             if (shootCommand)
             {
                 m_Shoot = true;
     public override void OnActionReceived(ActionBuffers actionBuffers)
 
     {
-        MoveAgent(actionBuffers.DiscreteActions);
+        MoveAgent(actionBuffers);
-        var discreteActionsOut = actionsOut.DiscreteActions;
-        discreteActionsOut[0] = 0;
-        discreteActionsOut[1] = 0;
-        discreteActionsOut[2] = 0;
+        var continuousActionsOut = actionsOut.ContinuousActions;
+        continuousActionsOut[0] = 0;
+        continuousActionsOut[1] = 0;
+        continuousActionsOut[2] = 0;
-            discreteActionsOut[2] = 2;
+            continuousActionsOut[2] = 1;
-            discreteActionsOut[0] = 1;
+            continuousActionsOut[0] = 1;
-            discreteActionsOut[2] = 1;
+            continuousActionsOut[2] = -1;
-            discreteActionsOut[0] = 2;
+            continuousActionsOut[0] = -1;
-        discreteActionsOut[3] = Input.GetKey(KeyCode.Space) ? 1 : 0;
+        var discreteActionsOut = actionsOut.DiscreteActions;
+        discreteActionsOut[0] = Input.GetKey(KeyCode.Space) ? 1 : 0;
     }
 
     public override void OnEpisodeBegin()

Project/Assets/ML-Agents/Examples/GridWorld/Prefabs/Area.prefab

   m_Name: 
   m_EditorClassIdentifier: 
   m_BrainParameters:
-    vectorObservationSize: 0
-    numStackedVectorObservations: 1
-    vectorActionSize: 05000000
-    vectorActionDescriptions: []
-    vectorActionSpaceType: 0
+    VectorObservationSize: 0
+    NumStackedVectorObservations: 1
+    m_ActionSpec:
+      m_NumContinuousActions: 0
+      BranchSizes: 05000000
+    VectorActionSize: 05000000
+    VectorActionDescriptions: []
+    VectorActionSpaceType: 0
+    hasUpgradedBrainParametersWithActionSpec: 1
-  m_InferenceDevice: 0
+  m_InferenceDevice: 2
+  m_UseChildActuators: 1
+  m_ObservableAttributeHandling: 0
 --- !u!114 &114650561397225712
 MonoBehaviour:
   m_ObjectHideFlags: 0
   agentParameters:
     maxStep: 0
   hasUpgradedFromAgentParameters: 1
-  maxStep: 100
+  MaxStep: 100
   area: {fileID: 114704252266302846}
   timeBetweenDecisionsAtInference: 0.15
   renderCamera: {fileID: 0}
   m_Width: 84
   m_Height: 64
   m_Grayscale: 0
+  m_ObservationStacks: 1
   m_Compression: 1
 --- !u!114 &7980686505185502968
 MonoBehaviour:
   m_Script: {fileID: 11500000, guid: 3a6da8f78a394c6ab027688eab81e04d, type: 3}
   m_Name: 
   m_EditorClassIdentifier: 
+  debugCommandLineOverride: 
 --- !u!1 &1625008366184734
 GameObject:
   m_ObjectHideFlags: 0
   trueAgent: {fileID: 1488387672112076}
   goalPref: {fileID: 1508142483324970, guid: 1ec4e4e96e7514d45b7ebc3ba5a9a481, type: 3}
   pitPref: {fileID: 1811317785436014, guid: d13ee2db77b3a4dcc8664d2fe2a0f219, type: 3}
+  numberOfObstacles: 1
 --- !u!1 &1656910849934022
 GameObject:
   m_ObjectHideFlags: 0

Project/Assets/ML-Agents/Examples/GridWorld/Scenes/GridWorld.unity

   m_ReflectionIntensity: 1
   m_CustomReflection: {fileID: 0}
   m_Sun: {fileID: 0}
-  m_IndirectSpecularColor: {r: 0.44971168, g: 0.4997775, b: 0.57563686, a: 1}
+  m_IndirectSpecularColor: {r: 0.4497121, g: 0.49977785, b: 0.57563704, a: 1}
   m_UseRadianceAmbientProbe: 0
 --- !u!157 &3
 LightmapSettings:
   agentParameters:
     maxStep: 100
   hasUpgradedFromAgentParameters: 1
-  maxStep: 100
+  MaxStep: 100
   area: {fileID: 1795599557}
   timeBetweenDecisionsAtInference: 0.15
   renderCamera: {fileID: 797520692}
   m_Name: 
   m_EditorClassIdentifier: 
   m_BrainParameters:
-    vectorObservationSize: 0
-    numStackedVectorObservations: 1
-    vectorActionSize: 05000000
-    vectorActionDescriptions: []
-    vectorActionSpaceType: 0
+    VectorObservationSize: 0
+    NumStackedVectorObservations: 1
+    m_ActionSpec:
+      m_NumContinuousActions: 0
+      BranchSizes: 05000000
+    VectorActionSize: 05000000
+    VectorActionDescriptions: []
+    VectorActionSpaceType: 0
+    hasUpgradedBrainParametersWithActionSpec: 1
-  m_InferenceDevice: 0
+  m_InferenceDevice: 2
+  m_UseChildActuators: 1
+  m_ObservableAttributeHandling: 0
 --- !u!114 &125487791
 MonoBehaviour:
   m_ObjectHideFlags: 0
   m_RenderTexture: {fileID: 8400000, guid: 114608d5384404f89bff4b6f88432958, type: 2}
   m_SensorName: RenderTextureSensor
   m_Grayscale: 0
+  m_ObservationStacks: 1
   m_Compression: 1
 --- !u!1 &260425459
 GameObject:
   trueAgent: {fileID: 125487785}
   goalPref: {fileID: 1508142483324970, guid: 1ec4e4e96e7514d45b7ebc3ba5a9a481, type: 3}
   pitPref: {fileID: 1811317785436014, guid: d13ee2db77b3a4dcc8664d2fe2a0f219, type: 3}
+  numberOfObstacles: 1
 --- !u!4 &1795599558
 Transform:
   m_ObjectHideFlags: 0

Project/Assets/ML-Agents/Examples/GridWorld/Scripts/GridAgent.cs

 using System.Linq;
 using Unity.MLAgents;
 using Unity.MLAgents.Actuators;
+using UnityEngine.Rendering;
 using UnityEngine.Serialization;
 
 public class GridAgent : Agent
         if (maskActions)
         {
             // Prevents the agent from picking an action that would make it collide with a wall
-            var positionX = (int)transform.position.x;
-            var positionZ = (int)transform.position.z;
+            var positionX = (int)transform.localPosition.x;
+            var positionZ = (int)transform.localPosition.z;
             var maxPosition = (int)m_ResetParams.GetWithDefault("gridSize", 5f) - 1;
 
             if (positionX == 0)
 
     void WaitTimeInference()
     {
-        if (renderCamera != null)
+        if (renderCamera != null && SystemInfo.graphicsDeviceType != GraphicsDeviceType.Null)
         {
             renderCamera.Render();
         }

Project/Assets/ML-Agents/Examples/GridWorld/Scripts/GridArea.cs

     public GameObject goalPref;
     public GameObject pitPref;
     GameObject[] m_Objects;
+    public int numberOfObstacles = 1;
 
     GameObject m_Plane;
     GameObject m_Sn;
         transform.position = m_InitialPosition * (m_ResetParams.GetWithDefault("gridSize", 5f) + 1);
         var playersList = new List<int>();
 
-        for (var i = 0; i < (int)m_ResetParams.GetWithDefault("numObstacles", 1); i++)
+        for (var i = 0; i < (int)m_ResetParams.GetWithDefault("numObstacles", numberOfObstacles); i++)
         {
             playersList.Add(1);
         }

Project/Assets/ML-Agents/Examples/Hallway/Prefabs/SymbolFinderArea.prefab

   m_BrainParameters:
     VectorObservationSize: 1
     NumStackedVectorObservations: 3
+    m_ActionSpec:
+      m_NumContinuousActions: 0
+      BranchSizes: 05000000
+    hasUpgradedBrainParametersWithActionSpec: 1
-  m_InferenceDevice: 0
+  m_InferenceDevice: 2
   m_BehaviorType: 0
   m_BehaviorName: Hallway
   TeamId: 0

Project/Assets/ML-Agents/Examples/Hallway/Prefabs/VisualSymbolFinderArea.prefab

   m_Name: 
   m_EditorClassIdentifier: 
   m_BrainParameters:
-    vectorObservationSize: 0
-    numStackedVectorObservations: 1
-    vectorActionSize: 05000000
-    vectorActionDescriptions: []
-    vectorActionSpaceType: 0
+    VectorObservationSize: 0
+    NumStackedVectorObservations: 1
+    m_ActionSpec:
+      m_NumContinuousActions: 0
+      BranchSizes: 05000000
+    VectorActionSize: 05000000
+    VectorActionDescriptions: []
+    VectorActionSpaceType: 0
+    hasUpgradedBrainParametersWithActionSpec: 1
-  m_InferenceDevice: 0
+  m_InferenceDevice: 2
-  m_TeamID: 0
-  m_useChildSensors: 1
+  TeamId: 0
+  m_UseChildSensors: 1
+  m_UseChildActuators: 1
+  m_ObservableAttributeHandling: 0
 --- !u!114 &114451776683649118
 MonoBehaviour:
   m_ObjectHideFlags: 0
   m_Script: {fileID: 11500000, guid: b446afae240924105b36d07e8d17a608, type: 3}
   m_Name: 
   m_EditorClassIdentifier: 
-  maxStep: 3000
+  agentParameters:
+    maxStep: 0
+  hasUpgradedFromAgentParameters: 1
+  MaxStep: 3000
   ground: {fileID: 1625056884785366}
   area: {fileID: 1689874756253538}
   symbolOGoal: {fileID: 1800868804754718}
   m_Script: {fileID: 11500000, guid: 282f342c2ab144bf38be65d4d0c4e07d, type: 3}
   m_Name: 
   m_EditorClassIdentifier: 
-  camera: {fileID: 20961984019151212}
-  sensorName: CameraSensor
-  width: 84
-  height: 84
-  grayscale: 0
-  compression: 1
+  m_Camera: {fileID: 20961984019151212}
+  m_SensorName: CameraSensor
+  m_Width: 84
+  m_Height: 84
+  m_Grayscale: 0
+  m_ObservationStacks: 1
+  m_Compression: 1
 --- !u!114 &640264344416331590
 MonoBehaviour:
   m_ObjectHideFlags: 0
   m_Name: 
   m_EditorClassIdentifier: 
   DecisionPeriod: 6
-  RepeatAction: 1
-  offsetStep: 0
+  TakeActionsBetweenDecisions: 1
 --- !u!1 &1377584197416466
 GameObject:
   m_ObjectHideFlags: 0

Project/Assets/ML-Agents/Examples/Hallway/Scripts/HallwayAgent.cs

     Renderer m_GroundRenderer;
     HallwaySettings m_HallwaySettings;
     int m_Selection;
+    StatsRecorder m_statsRecorder;
 
     public override void Initialize()
     {
         m_GroundMaterial = m_GroundRenderer.material;
+        m_statsRecorder = Academy.Instance.StatsRecorder;
     }
 
     public override void CollectObservations(VectorSensor sensor)
             {
                 SetReward(1f);
                 StartCoroutine(GoalScoredSwapGroundMaterial(m_HallwaySettings.goalScoredMaterial, 0.5f));
+                m_statsRecorder.Add("Goal/Correct", 1, StatAggregationMethod.Sum);
+                m_statsRecorder.Add("Goal/Wrong", 1, StatAggregationMethod.Sum);
             }
             EndEpisode();
         }
             symbolXGoal.transform.position = new Vector3(7f, 0.5f, 22.29f) + area.transform.position;
             symbolOGoal.transform.position = new Vector3(-7f, 0.5f, 22.29f) + area.transform.position;
         }
+        m_statsRecorder.Add("Goal/Correct", 0, StatAggregationMethod.Sum);
+        m_statsRecorder.Add("Goal/Wrong", 0, StatAggregationMethod.Sum);
     }
 }

Project/Assets/ML-Agents/Examples/PushBlock/Prefabs/PushBlockArea.prefab

   m_Name: 
   m_EditorClassIdentifier: 
   m_BrainParameters:
-    vectorObservationSize: 0
-    numStackedVectorObservations: 2
-    vectorActionSize: 07000000
-    vectorActionDescriptions: []
-    vectorActionSpaceType: 0
+    VectorObservationSize: 0
+    NumStackedVectorObservations: 2
+    m_ActionSpec:
+      m_NumContinuousActions: 0
+      BranchSizes: 07000000
+    VectorActionSize: 07000000
+    VectorActionDescriptions: []
+    VectorActionSpaceType: 0
+    hasUpgradedBrainParametersWithActionSpec: 1
-  m_InferenceDevice: 0
+  m_InferenceDevice: 2
+  m_UseChildActuators: 1
+  m_ObservableAttributeHandling: 0
 --- !u!114 &114505490781873732
 MonoBehaviour:
   m_ObjectHideFlags: 0
   agentParameters:
     maxStep: 0
   hasUpgradedFromAgentParameters: 1
-  maxStep: 5000
+  MaxStep: 5000
   ground: {fileID: 1500989011945850}
   area: {fileID: 1125452240183160}
   areaBounds:
   m_EditorClassIdentifier: 
   DecisionPeriod: 5
   TakeActionsBetweenDecisions: 1
-  offsetStep: 0
 --- !u!114 &4081319787948195948
 MonoBehaviour:
   m_ObjectHideFlags: 0
   m_Script: {fileID: 11500000, guid: 3a6da8f78a394c6ab027688eab81e04d, type: 3}
   m_Name: 
   m_EditorClassIdentifier: 
+  debugCommandLineOverride: 
 --- !u!1 &1500989011945850
 GameObject:
   m_ObjectHideFlags: 0

Project/Assets/ML-Agents/Examples/PushBlock/Prefabs/PushBlockVisualArea.prefab

   m_BrainParameters:
     VectorObservationSize: 0
     NumStackedVectorObservations: 1
+    m_ActionSpec:
+      m_NumContinuousActions: 0
+      BranchSizes: 07000000
+    hasUpgradedBrainParametersWithActionSpec: 1
-  m_InferenceDevice: 0
+  m_InferenceDevice: 2
+  m_UseChildActuators: 1
   m_ObservableAttributeHandling: 0
 --- !u!114 &114812843792483960
 MonoBehaviour:
   m_Width: 84
   m_Height: 84
   m_Grayscale: 0
+  m_ObservationStacks: 1
   m_Compression: 1
 --- !u!114 &9049837659352187721
 MonoBehaviour:

Project/Assets/ML-Agents/Examples/Pyramids/Prefabs/AreaPB.prefab

   m_Name: 
   m_EditorClassIdentifier: 
   m_BrainParameters:
-    vectorObservationSize: 4
-    numStackedVectorObservations: 1
-    vectorActionSize: 05000000
-    vectorActionDescriptions: []
-    vectorActionSpaceType: 0
+    VectorObservationSize: 4
+    NumStackedVectorObservations: 1
+    m_ActionSpec:
+      m_NumContinuousActions: 0
+      BranchSizes: 05000000
+    VectorActionSize: 05000000
+    VectorActionDescriptions: []
+    VectorActionSpaceType: 0
+    hasUpgradedBrainParametersWithActionSpec: 1
-  m_InferenceDevice: 0
+  m_InferenceDevice: 2
+  m_UseChildActuators: 1
+  m_ObservableAttributeHandling: 0
 --- !u!114 &114937736047215868
 MonoBehaviour:
   m_ObjectHideFlags: 0
   agentParameters:
     maxStep: 0
   hasUpgradedFromAgentParameters: 1
-  maxStep: 5000
+  MaxStep: 5000
   area: {fileID: 1464170487903594}
   areaSwitch: {fileID: 1432086782037750}
   useVectorObs: 1
   m_EditorClassIdentifier: 
   DecisionPeriod: 5
   TakeActionsBetweenDecisions: 1
-  offsetStep: 0
 --- !u!114 &5712624269609438939
 MonoBehaviour:
   m_ObjectHideFlags: 0
   m_Script: {fileID: 11500000, guid: 3a6da8f78a394c6ab027688eab81e04d, type: 3}
   m_Name: 
   m_EditorClassIdentifier: 
+  debugCommandLineOverride: 
 --- !u!1 &1148882946833254
 GameObject:
   m_ObjectHideFlags: 0

Project/Assets/ML-Agents/Examples/Pyramids/Prefabs/VisualAreaPyramids.prefab

   m_Name: 
   m_EditorClassIdentifier: 
   m_BrainParameters:
-    vectorObservationSize: 0
-    numStackedVectorObservations: 1
-    vectorActionSize: 05000000
-    vectorActionDescriptions: []
-    vectorActionSpaceType: 0
+    VectorObservationSize: 0
+    NumStackedVectorObservations: 1
+    m_ActionSpec:
+      m_NumContinuousActions: 0
+      BranchSizes: 05000000
+    VectorActionSize: 05000000
+    VectorActionDescriptions: []
+    VectorActionSpaceType: 0
+    hasUpgradedBrainParametersWithActionSpec: 1
-  m_InferenceDevice: 0
+  m_InferenceDevice: 2
-  m_TeamID: 0
-  m_useChildSensors: 1
+  TeamId: 0
+  m_UseChildSensors: 1
+  m_UseChildActuators: 1
+  m_ObservableAttributeHandling: 0
 --- !u!114 &114741503533626942
 MonoBehaviour:
   m_ObjectHideFlags: 0
   m_Script: {fileID: 11500000, guid: b8db44472779248d3be46895c4d562d5, type: 3}
   m_Name: 
   m_EditorClassIdentifier: 
-  maxStep: 5000
+  agentParameters:
+    maxStep: 0
+  hasUpgradedFromAgentParameters: 1
+  MaxStep: 5000
   area: {fileID: 1055559745433172}
   areaSwitch: {fileID: 1212218760704844}
   useVectorObs: 0
   m_Script: {fileID: 11500000, guid: 282f342c2ab144bf38be65d4d0c4e07d, type: 3}
   m_Name: 
   m_EditorClassIdentifier: 
-  camera: {fileID: 20712684238256298}
-  sensorName: CameraSensor
-  width: 84
-  height: 84
-  grayscale: 0
-  compression: 1
+  m_Camera: {fileID: 20712684238256298}
+  m_SensorName: CameraSensor
+  m_Width: 84
+  m_Height: 84
+  m_Grayscale: 0
+  m_ObservationStacks: 1
+  m_Compression: 1
 --- !u!114 &9216598927300453297
 MonoBehaviour:
   m_ObjectHideFlags: 0
   m_Name: 
   m_EditorClassIdentifier: 
   DecisionPeriod: 5
-  RepeatAction: 1
-  offsetStep: 0
+  TakeActionsBetweenDecisions: 1
 --- !u!1 &1747856067778386
 GameObject:
   m_ObjectHideFlags: 0

Project/Assets/ML-Agents/Examples/Reacher/Prefabs/Agent.prefab

   m_Name: 
   m_EditorClassIdentifier: 
   m_BrainParameters:
-    vectorObservationSize: 33
-    numStackedVectorObservations: 1
-    vectorActionSize: 04000000
-    vectorActionDescriptions: []
-    vectorActionSpaceType: 1
+    VectorObservationSize: 33
+    NumStackedVectorObservations: 1
+    m_ActionSpec:
+      m_NumContinuousActions: 4
+      BranchSizes: 
+    VectorActionSize: 04000000
+    VectorActionDescriptions: []
+    VectorActionSpaceType: 1
+    hasUpgradedBrainParametersWithActionSpec: 1
-  m_InferenceDevice: 0
+  m_InferenceDevice: 2
+  m_UseChildActuators: 1
+  m_ObservableAttributeHandling: 0
 --- !u!114 &114955921823023820
 MonoBehaviour:
   m_ObjectHideFlags: 0
   agentParameters:
     maxStep: 0
   hasUpgradedFromAgentParameters: 1
-  maxStep: 4000
+  MaxStep: 4000
   pendulumA: {fileID: 1644872085946016}
   pendulumB: {fileID: 1053261483945176}
   hand: {fileID: 1654288206095398}
   m_EditorClassIdentifier: 
   DecisionPeriod: 4
   TakeActionsBetweenDecisions: 1
-  offsetStep: 0
 --- !u!114 &7840105453417110232
 MonoBehaviour:
   m_ObjectHideFlags: 0
   m_Script: {fileID: 11500000, guid: 3a6da8f78a394c6ab027688eab81e04d, type: 3}
   m_Name: 
   m_EditorClassIdentifier: 
+  debugCommandLineOverride: 
 --- !u!1 &1644872085946016
 GameObject:
   m_ObjectHideFlags: 0

Project/Assets/ML-Agents/Examples/Reacher/Scripts/ReacherAgent.cs

     /// The agent's four actions correspond to torques on each of the two joints.
     /// </summary>
     public override void OnActionReceived(ActionBuffers actionBuffers)
-
     {
         m_GoalDegree += m_GoalSpeed;
         UpdateGoalPosition();

Project/Assets/ML-Agents/Examples/SharedAssets/Materials/Purple.mat

 Material:
   serializedVersion: 6
   m_ObjectHideFlags: 0
-  m_PrefabParentObject: {fileID: 0}
-  m_PrefabInternal: {fileID: 0}
+  m_CorrespondingSourceObject: {fileID: 0}
+  m_PrefabInstance: {fileID: 0}
+  m_PrefabAsset: {fileID: 0}
   m_Name: Purple
   m_Shader: {fileID: 46, guid: 0000000000000000f000000000000000, type: 0}
   m_ShaderKeywords: _GLOSSYREFLECTIONS_OFF _SPECULARHIGHLIGHTS_OFF

Project/Assets/ML-Agents/Examples/SharedAssets/Scripts/ModelOverrider.cs

         const string k_CommandLineModelOverrideDirectoryFlag = "--mlagents-override-model-directory";
         const string k_CommandLineModelOverrideExtensionFlag = "--mlagents-override-model-extension";
         const string k_CommandLineQuitAfterEpisodesFlag = "--mlagents-quit-after-episodes";
+        const string k_CommandLineQuitAfterSeconds = "--mlagents-quit-after-seconds";
         const string k_CommandLineQuitOnLoadFailure = "--mlagents-quit-on-load-failure";
 
         // The attached Agent
         // Max episodes to run. Only used if > 0
         // Will default to 1 if override models are specified, otherwise 0.
         int m_MaxEpisodes;
+
+        // Deadline - exit if the time exceeds this
+        DateTime m_Deadline = DateTime.MaxValue;
 
         int m_NumSteps;
         int m_PreviousNumSteps;
         void GetAssetPathFromCommandLine()
         {
             var maxEpisodes = 0;
+            var timeoutSeconds = 0;
+
             string[] commandLineArgsOverride = null;
             if (!string.IsNullOrEmpty(debugCommandLineOverride) && Application.isEditor)
             {
                 {
                     Int32.TryParse(args[i + 1], out maxEpisodes);
                 }
+                else if (args[i] == k_CommandLineQuitAfterSeconds && i < args.Length - 1)
+                {
+                    Int32.TryParse(args[i + 1], out timeoutSeconds);
+                }
                 else if (args[i] == k_CommandLineQuitOnLoadFailure)
                 {
                     m_QuitOnLoadFailure = true;
                 m_MaxEpisodes = maxEpisodes > 0 ? maxEpisodes : 1;
                 Debug.Log($"setting m_MaxEpisodes to {maxEpisodes}");
             }
+
+            if (timeoutSeconds > 0)
+            {
+                m_Deadline = DateTime.Now + TimeSpan.FromSeconds(timeoutSeconds);
+                Debug.Log($"setting deadline to {timeoutSeconds} from now.");
+
+            }
         }
 
         void OnEnable()
                     EditorApplication.isPlaying = false;
 #endif
                 }
+                else if (DateTime.Now >= m_Deadline)
+                {
+                    Debug.Log(
+                        $"Deadline exceeded. " +
+                        $"{TotalCompletedEpisodes}/{m_MaxEpisodes} episodes and " +
+                        $"{TotalNumSteps}/{m_MaxEpisodes * m_Agent.MaxStep} steps completed. Exiting.");
+                    Application.Quit(0);
+#if UNITY_EDITOR
+                    EditorApplication.isPlaying = false;
+#endif
+                }
+
             m_NumSteps++;
         }
 

Project/Assets/ML-Agents/Examples/SharedAssets/Scripts/SensorBase.cs

             float[] buffer = new float[numFloats];
             WriteObservation(buffer);
 
-            writer.AddRange(buffer);
+            writer.AddList(buffer);
 
             return numFloats;
         }

Project/Assets/ML-Agents/Examples/Soccer/Scenes/SoccerTwos.unity

   m_Modification:
     m_TransformParent: {fileID: 0}
     m_Modifications:
+    - target: {fileID: 1141134673700168, guid: 54f3340298537426e96a6cc530e2d5d8, type: 3}
+      propertyPath: m_Name
+      value: SoccerFieldTwos
+      objectReference: {fileID: 0}
     - target: {fileID: 4558743310993102, guid: 54f3340298537426e96a6cc530e2d5d8, type: 3}
       propertyPath: m_LocalPosition.x
       value: 0
     - target: {fileID: 4558743310993102, guid: 54f3340298537426e96a6cc530e2d5d8, type: 3}
       propertyPath: m_RootOrder
       value: 4
+      objectReference: {fileID: 0}
+    - target: {fileID: 4558743310993102, guid: 54f3340298537426e96a6cc530e2d5d8, type: 3}
+      propertyPath: m_LocalEulerAnglesHint.x
+      value: 0
+      objectReference: {fileID: 0}
+    - target: {fileID: 4558743310993102, guid: 54f3340298537426e96a6cc530e2d5d8, type: 3}
+      propertyPath: m_LocalEulerAnglesHint.y
+      value: 0
+      objectReference: {fileID: 0}
+    - target: {fileID: 4558743310993102, guid: 54f3340298537426e96a6cc530e2d5d8, type: 3}
+      propertyPath: m_LocalEulerAnglesHint.z
+      value: 0
       objectReference: {fileID: 0}
     m_RemovedComponents: []
   m_SourcePrefab: {fileID: 100100000, guid: 54f3340298537426e96a6cc530e2d5d8, type: 3}
   m_Name: 
   m_EditorClassIdentifier: 
   gravityMultiplier: 1
-  monitorVerticalOffset: 0
+  reuseCollisionCallbacks: 1
 --- !u!114 &1574236051
 MonoBehaviour:
   m_ObjectHideFlags: 0
   blueMaterial: {fileID: 2100000, guid: c9fa44c2c3f8ce74ca39a3355ea42631, type: 2}
   randomizePlayersTeamForTraining: 0
   agentRunSpeed: 2
-  strikerPunish: -0.1
-  strikerReward: 1
-  goaliePunish: -1
-  goalieReward: 0.1
 --- !u!1001 &1606160104
 PrefabInstance:
   m_ObjectHideFlags: 0

Project/Assets/ML-Agents/Examples/Soccer/Scenes/StrikersVsGoalie.unity

   maximumDeltaTime: 0.33333334
   solverIterations: 6
   solverVelocityIterations: 1
+  reuseCollisionCallbacks: 1
 --- !u!114 &1574236051
 MonoBehaviour:
   m_ObjectHideFlags: 0

Project/Assets/ML-Agents/Examples/Soccer/Scripts/AgentSoccer.cs

-using System;
 using UnityEngine;
 using Unity.MLAgents;
 using Unity.MLAgents.Actuators;

Project/Assets/ML-Agents/Examples/Template/Scripts/TemplateAgent.cs

-using UnityEngine;
 using Unity.MLAgents;
 using Unity.MLAgents.Actuators;
 using Unity.MLAgents.Sensors;

Project/Assets/ML-Agents/Examples/Tennis/Prefabs/TennisArea.prefab

   m_BrainParameters:
     VectorObservationSize: 9
     NumStackedVectorObservations: 3
+    m_ActionSpec:
+      m_NumContinuousActions: 3
+      BranchSizes: 
+    hasUpgradedBrainParametersWithActionSpec: 1
-  m_InferenceDevice: 0
+  m_InferenceDevice: 2
+  m_UseChildActuators: 1
+  m_ObservableAttributeHandling: 0
 --- !u!114 &114915946461826994
 MonoBehaviour:
   m_ObjectHideFlags: 0
   m_Script: {fileID: 11500000, guid: 3a6da8f78a394c6ab027688eab81e04d, type: 3}
   m_Name: 
   m_EditorClassIdentifier: 
+  debugCommandLineOverride: 
 --- !u!1 &1194790474478638
 GameObject:
   m_ObjectHideFlags: 0
   m_BrainParameters:
     VectorObservationSize: 9
     NumStackedVectorObservations: 3
+    m_ActionSpec:
+      m_NumContinuousActions: 3
+      BranchSizes: 
+    hasUpgradedBrainParametersWithActionSpec: 1
-  m_InferenceDevice: 0
+  m_InferenceDevice: 2
+  m_UseChildActuators: 1
+  m_ObservableAttributeHandling: 0
 --- !u!114 &114800310164848628
 MonoBehaviour:
   m_ObjectHideFlags: 0
   m_Script: {fileID: 11500000, guid: 3a6da8f78a394c6ab027688eab81e04d, type: 3}
   m_Name: 
   m_EditorClassIdentifier: 
+  debugCommandLineOverride: 
 --- !u!1 &1969551055586186
 GameObject:
   m_ObjectHideFlags: 0

Project/Assets/ML-Agents/Examples/Walker/Prefabs/Ragdoll/WalkerRagdollBase.prefab

   m_BrainParameters:
     VectorObservationSize: 243
     NumStackedVectorObservations: 1
+    m_ActionSpec:
+      m_NumContinuousActions: 39
+      BranchSizes: 
+    hasUpgradedBrainParametersWithActionSpec: 1
-  m_InferenceDevice: 0
+  m_InferenceDevice: 2
+  m_UseChildActuators: 1
   m_ObservableAttributeHandling: 0
 --- !u!114 &7408209125961349353
 MonoBehaviour:
     maxStep: 0
   hasUpgradedFromAgentParameters: 1
   MaxStep: 5000
-  targetWalkingSpeed: 10
+  m_TargetWalkingSpeed: 10
-  walkDirectionMethod: 0
-  worldDirToWalk: {x: 1, y: 0, z: 0}
-  worldPosToWalkTo: {x: 0, y: 0, z: 0}
   target: {fileID: 0}
   hips: {fileID: 895268871264836332}
   chest: {fileID: 7933235354845945071}

Project/Assets/ML-Agents/Examples/Walker/Prefabs/Ragdoll/WalkerRagdollDySingleSpeedVariant.prefab

       value: 
       objectReference: {fileID: 11400000, guid: 47e7c480450ec4dcd9e4a04124e14ed4,
         type: 3}
+    - target: {fileID: 895268871377934297, guid: 765582efd9dda46ed98564603316353f,
+        type: 3}
+      propertyPath: m_InferenceDevice
+      value: 2
+      objectReference: {fileID: 0}
     - target: {fileID: 895268871377934298, guid: 765582efd9dda46ed98564603316353f,
         type: 3}
       propertyPath: m_LocalPosition.x

Project/Assets/ML-Agents/Examples/Walker/Scripts/WalkerAgent.cs

     const float m_maxWalkingSpeed = 10; //The max walking speed
 
     //Should the agent sample a new goal velocity each episode?
-    //If true, walkSpeed will be randomly set between zero and m_maxWalkingSpeed in OnEpisodeBegin() 
+    //If true, walkSpeed will be randomly set between zero and m_maxWalkingSpeed in OnEpisodeBegin()
     //If false, the goal velocity will be walkingSpeed
     public bool randomizeWalkSpeedEachEpisode;
 
         //ragdoll's avg vel
         var avgVel = GetAvgVelocity();
 
-        //current ragdoll velocity. normalized 
+        //current ragdoll velocity. normalized
         sensor.AddObservation(Vector3.Distance(velGoal, avgVel));
         //avg body vel relative to cube
         sensor.AddObservation(m_OrientationCube.transform.InverseTransformDirection(avgVel));
         Vector3 avgVel = Vector3.zero;
 
         //ALL RBS
-        int numOfRB = 0;
+        int numOfRb = 0;
-            numOfRB++;
+            numOfRb++;
-        avgVel = velSum / numOfRB;
+        avgVel = velSum / numOfRb;
         return avgVel;
     }
 

Project/Assets/ML-Agents/Examples/WallJump/Prefabs/WallJumpArea.prefab

   m_BrainParameters:
     VectorObservationSize: 4
     NumStackedVectorObservations: 6
+    m_ActionSpec:
+      m_NumContinuousActions: 0
+      BranchSizes: 03000000030000000300000002000000
+    hasUpgradedBrainParametersWithActionSpec: 1
-  m_InferenceDevice: 0
+  m_InferenceDevice: 2
+  m_UseChildActuators: 1
   m_ObservableAttributeHandling: 0
 --- !u!114 &114925928594762506
 MonoBehaviour:

Project/Assets/ML-Agents/Examples/Worm/Prefabs/WormBasePrefab.prefab

   m_BrainParameters:
     VectorObservationSize: 64
     NumStackedVectorObservations: 1
+    m_ActionSpec:
+      m_NumContinuousActions: 9
+      BranchSizes: 
+    hasUpgradedBrainParametersWithActionSpec: 1
-  m_InferenceDevice: 0
+  m_InferenceDevice: 2
   m_BehaviorType: 0
   m_BehaviorName: WormDynamic
   TeamId: 0

Project/Assets/ML-Agents/TestScenes/TestCompressedTexture/TestTextureSensor.cs

 using UnityEngine;
 using Unity.MLAgents.Sensors;
-using Unity.MLAgents;
 
 public class TestTextureSensor : ISensor
 {

Project/Assets/ML-Agents/TestScenes/TestCompressedTexture/TestTextureSensorComponent.cs

 using UnityEngine;
-using UnityEngine.Serialization;
 using Unity.MLAgents.Sensors;
 
 

Project/ProjectSettings/UnityConnectSettings.asset

 UnityConnectSettings:
   m_ObjectHideFlags: 0
   serializedVersion: 1
-  m_Enabled: 1
+  m_Enabled: 0
   m_TestMode: 0
   m_EventOldUrl: https://api.uca.cloud.unity3d.com/v1/events
   m_EventUrl: https://cdp.cloud.unity3d.com/v1/events

README.md

 
 # Unity ML-Agents Toolkit
 
-[![docs badge](https://img.shields.io/badge/docs-reference-blue.svg)](https://github.com/Unity-Technologies/ml-agents/tree/release_8_docs/docs/)
+[![docs badge](https://img.shields.io/badge/docs-reference-blue.svg)](https://github.com/Unity-Technologies/ml-agents/tree/release_12_docs/docs/)
 
 [![license badge](https://img.shields.io/badge/license-Apache--2.0-green.svg)](LICENSE)
 
 
 ## Features
 
-- 15+ [example Unity environments](docs/Learning-Environment-Examples.md)
+- 18+ [example Unity environments](docs/Learning-Environment-Examples.md)
-- Built-in support for Imitation Learning through Behavioral Cloning or
-  Generative Adversarial Imitation Learning
+- Built-in support for Imitation Learning through Behavioral Cloning (BC) or
+  Generative Adversarial Imitation Learning (GAIL)
 - Self-play mechanism for training agents in adversarial scenarios
 - Easily definable Curriculum Learning scenarios for complex tasks
 - Train robust agents using environment randomization
 
 ## Releases & Documentation
 
-**Our latest, stable release is `Release 8`. Click
-[here](https://github.com/Unity-Technologies/ml-agents/tree/release_8_docs/docs/Readme.md)
+**Our latest, stable release is `Release 12`. Click
+[here](https://github.com/Unity-Technologies/ml-agents/tree/release_12_docs/docs/Readme.md)
 to get started with the latest release of ML-Agents.**
 
 The table below lists all our releases, including our `master` branch which is
 - The **Documentation** links in the table below include installation and usage
   instructions specific to each release. Remember to always use the
   documentation that corresponds to the release version you're using.
+- The `com.unity.ml-agents` package is [verified](https://docs.unity3d.com/2020.1/Documentation/Manual/pack-safe.html)
+  for Unity 2020.1 and later. Verified packages releases are numbered 1.0.x.
-| **Version** | **Release Date** | **Source** | **Documentation** | **Download** |
-|:-------:|:------:|:-------------:|:-------:|:------------:|
-| **master (unstable)** | -- | [source](https://github.com/Unity-Technologies/ml-agents/tree/master) | [docs](https://github.com/Unity-Technologies/ml-agents/tree/master/docs/Readme.md) | [download](https://github.com/Unity-Technologies/ml-agents/archive/master.zip) |
-| **Release 8** | **October 14, 2020** | **[source](https://github.com/Unity-Technologies/ml-agents/tree/release_8)** | **[docs](https://github.com/Unity-Technologies/ml-agents/tree/release_8_docs/docs/Readme.md)** | **[download](https://github.com/Unity-Technologies/ml-agents/archive/release_8.zip)** |
-| **Release 7** | September 16, 2020 | [source](https://github.com/Unity-Technologies/ml-agents/tree/release_7) | [docs](https://github.com/Unity-Technologies/ml-agents/tree/release_7_docs/docs/Readme.md) | [download](https://github.com/Unity-Technologies/ml-agents/archive/release_7.zip) |
-| **Release 6** | August 12, 2020 | [source](https://github.com/Unity-Technologies/ml-agents/tree/release_6) | [docs](https://github.com/Unity-Technologies/ml-agents/tree/release_6_docs/docs/Readme.md) | [download](https://github.com/Unity-Technologies/ml-agents/archive/release_6.zip) |
-| **Release 5** | July 31, 2020 | [source](https://github.com/Unity-Technologies/ml-agents/tree/release_5) | [docs](https://github.com/Unity-Technologies/ml-agents/tree/release_5_docs/docs/Readme.md) | [download](https://github.com/Unity-Technologies/ml-agents/archive/release_5.zip) |
-| **Release 4** | July 15, 2020 | [source](https://github.com/Unity-Technologies/ml-agents/tree/release_4) | [docs](https://github.com/Unity-Technologies/ml-agents/tree/release_4_docs/docs/Readme.md) | [download](https://github.com/Unity-Technologies/ml-agents/archive/release_4.zip) |
-| **Release 3** | June 10, 2020 | [source](https://github.com/Unity-Technologies/ml-agents/tree/release_3) | [docs](https://github.com/Unity-Technologies/ml-agents/tree/release_3_docs/docs/Readme.md) | [download](https://github.com/Unity-Technologies/ml-agents/archive/release_3.zip) |
-| **Release 2** | May 20, 2020 | [source](https://github.com/Unity-Technologies/ml-agents/tree/release_2) | [docs](https://github.com/Unity-Technologies/ml-agents/tree/release_2_docs/docs/Readme.md) | [download](https://github.com/Unity-Technologies/ml-agents/archive/release_2.zip) |
-
-## Citation
+| **Version** | **Release Date** | **Source** | **Documentation** | **Download** | **Python Package** | **Unity Package** |
+|:-------:|:------:|:-------------:|:-------:|:------------:|:------------:|:------------:|
+| **master (unstable)** | -- | [source](https://github.com/Unity-Technologies/ml-agents/tree/master) | [docs](https://github.com/Unity-Technologies/ml-agents/tree/master/docs/Readme.md) | [download](https://github.com/Unity-Technologies/ml-agents/archive/master.zip) | -- | -- |
+| **Release 12** | **December 22, 2020** | **[source](https://github.com/Unity-Technologies/ml-agents/tree/release_12)** | **[docs](https://github.com/Unity-Technologies/ml-agents/tree/release_12_docs/docs/Readme.md)** | **[download](https://github.com/Unity-Technologies/ml-agents/archive/release_12.zip)** | **[0.23.0](https://pypi.org/project/mlagents/0.23.0/)** | **[1.7.2](https://docs.unity3d.com/Packages/com.unity.ml-agents@1.7/manual/index.html)** |
+| **Release 11** | December 21, 2020 | [source](https://github.com/Unity-Technologies/ml-agents/tree/release_11) | [docs](https://github.com/Unity-Technologies/ml-agents/tree/release_11_docs/docs/Readme.md) | [download](https://github.com/Unity-Technologies/ml-agents/archive/release_11.zip) | [0.23.0](https://pypi.org/project/mlagents/0.23.0/) | [1.7.0](https://docs.unity3d.com/Packages/com.unity.ml-agents@1.7/manual/index.html) |
+| **Release 10** | November 18, 2020 | [source](https://github.com/Unity-Technologies/ml-agents/tree/release_10) | [docs](https://github.com/Unity-Technologies/ml-agents/tree/release_10_docs/docs/Readme.md) | [download](https://github.com/Unity-Technologies/ml-agents/archive/release_10.zip) | [0.22.0](https://pypi.org/project/mlagents/0.22.0/) | [1.6.0](https://docs.unity3d.com/Packages/com.unity.ml-agents@1.6/manual/index.html) |
+| **Verified Package 1.0.6** | **November 16, 2020** | **[source](https://github.com/Unity-Technologies/ml-agents/tree/com.unity.ml-agents_1.0.6)** | **[docs](https://github.com/Unity-Technologies/ml-agents/blob/release_2_verified_docs/docs/Readme.md)** | **[download](https://github.com/Unity-Technologies/ml-agents/archive/com.unity.ml-agents_1.0.6.zip)** | **[0.16.1](https://pypi.org/project/mlagents/0.16.1/)** | **[1.0.6](https://docs.unity3d.com/Packages/com.unity.ml-agents@1.0/manual/index.html)** |
+| **Release 9** | November 4, 2020 | [source](https://github.com/Unity-Technologies/ml-agents/tree/release_9) | [docs](https://github.com/Unity-Technologies/ml-agents/tree/release_9_docs/docs/Readme.md) | [download](https://github.com/Unity-Technologies/ml-agents/archive/release_9.zip) | [0.21.1](https://pypi.org/project/mlagents/0.21.1/) | [1.5.0](https://docs.unity3d.com/Packages/com.unity.ml-agents@1.5/manual/index.html) |
+| **Release 8** | October 14, 2020 | [source](https://github.com/Unity-Technologies/ml-agents/tree/release_8) | [docs](https://github.com/Unity-Technologies/ml-agents/tree/release_8_docs/docs/Readme.md) | [download](https://github.com/Unity-Technologies/ml-agents/archive/release_8.zip) | [0.21.0](https://pypi.org/project/mlagents/0.21.0/) | [1.5.0](https://docs.unity3d.com/Packages/com.unity.ml-agents@1.5/manual/index.html) |
+| **Verified Package 1.0.5** | September 23, 2020 | [source](https://github.com/Unity-Technologies/ml-agents/tree/com.unity.ml-agents_1.0.5) | [docs](https://github.com/Unity-Technologies/ml-agents/blob/release_2_verified_docs/docs/Readme.md) | [download](https://github.com/Unity-Technologies/ml-agents/archive/com.unity.ml-agents_1.0.5.zip) | [0.16.1](https://pypi.org/project/mlagents/0.16.1/) | [1.0.5](https://docs.unity3d.com/Packages/com.unity.ml-agents@1.0/manual/index.html) |
+| **Release 7** | September 16, 2020 | [source](https://github.com/Unity-Technologies/ml-agents/tree/release_7) | [docs](https://github.com/Unity-Technologies/ml-agents/tree/release_7_docs/docs/Readme.md) | [download](https://github.com/Unity-Technologies/ml-agents/archive/release_7.zip) | [0.20.0](https://pypi.org/project/mlagents/0.20.0/) | [1.4.0](https://docs.unity3d.com/Packages/com.unity.ml-agents@1.4/manual/index.html) |
 
 If you are a researcher interested in a discussion of Unity as an AI platform,
 see a pre-print of our
 
 ## Additional Resources
 
-We have published a series of blog posts that are relevant for ML-Agents:
+We have a Unity Learn course,
+[ML-Agents: Hummingsbird](https://learn.unity.com/course/ml-agents-hummingbirds),
+that provides a gentle introduction to Unity and the ML-Agents Toolkit.
+We've also partnered with
+[CodeMonkeyUnity](https://www.youtube.com/c/CodeMonkeyUnity) to create a
+[series of tutorial videos](https://www.youtube.com/playlist?list=PLzDRvYVwl53vehwiN_odYJkPBzcqFw110)
+on how to implement and use the ML-Agents Toolkit.
+
+We have also published a series of blog posts that are relevant for ML-Agents:
+
+- (December 28, 2020)
+  [Happy holidays from the Unity ML-Agents team!](https://blogs.unity3d.com/2020/12/28/happy-holidays-from-the-unity-ml-agents-team/)
+- (November 20, 2020)
+  [How Eidos-Montréal created Grid Sensors to improve observations for training agents](https://blogs.unity3d.com/2020/11/20/how-eidos-montreal-created-grid-sensors-to-improve-observations-for-training-agents/)
+- (November 11, 2020)
+  [2020 AI@Unity interns shoutout](https://blogs.unity3d.com/2020/11/11/2020-aiunity-interns-shoutout/)
 - (May 12, 2020)
   [Announcing ML-Agents Unity Package v1.0!](https://blogs.unity3d.com/2020/05/12/announcing-ml-agents-unity-package-v1-0/)
 - (February 28, 2020)
   and
   [Q-learning](https://blogs.unity3d.com/2017/08/22/unity-ai-reinforcement-learning-with-q-learning/))
 
-In addition to our own documentation, here are some additional, relevant
-articles:
-
-- [A Game Developer Learns Machine Learning](https://mikecann.co.uk/machine-learning/a-game-developer-learns-machine-learning-intent/)
-- [Explore Unity Technologies ML-Agents Exclusively on Intel Architecture](https://software.intel.com/en-us/articles/explore-unity-technologies-ml-agents-exclusively-on-intel-architecture)
-- [ML-Agents Penguins tutorial](https://learn.unity.com/project/ml-agents-penguins)
-
 ## Community and Feedback
 
 The ML-Agents Toolkit is an open-source project and we encourage and welcome
 
 For any other questions or feedback, connect directly with the ML-Agents team at
 ml-agents@unity3d.com.
+
+## Privacy
+
+In order to improve the developer experience for Unity ML-Agents Toolkit, we have added in-editor analytics.
+Please refer to "Information that is passively collected by Unity" in the
+[Unity Privacy Policy](https://unity3d.com/legal/privacy-policy).
 
 ## License
 

com.unity.ml-agents.extensions/Documentation~/Grid-Sensor.md

 # Summary
 
-The Grid Sensor combines the generality of data extraction from Raycasts with the image processing power of Convolutional Neural Networks. The Grid Sensor can be used to collect data in the general form of a "Width x Height x Channel" matrix which can be used for training Reinforcement Learning agents or for data analysis.
+The Grid Sensor is an alternative method for collecting observations which combines the generality of data extraction from Raycasts with the image processing power of Convolutional Neural Networks. The Grid Sensor can be used to collect data in the general form of a "Width x Height x Channel" matrix which can be used for training agent policies or for data analysis.
-In MLAgents there are 2 main sensors for observing information that is "physically" around the agent.
+In ML-Agents there are two main sensors for observing information that is "physically" around the agent.
+
-This is simple to implement and provides enough information for most simple games. When few are used, they are computationally fast. However, there are multiple limiting factors:
-* The rays need to be at the same height as the things the agent should observe
-* Objects can remain hidden by line of sight and if the knowledge of those objects is crucial to the success of the agent, then this limitation must be compensated for by the agents networks capacity (i.e., need a bigger brain with memory)
+Raycasts are simple to implement and provides enough information for most simple games. When few are used, they are also computationally lightweight. However, there are multiple limiting factors:
+
+* The rays need to be at the same height as the things the agent should observe.
+* Objects can remain hidden by line of sight and if the knowledge of those objects is crucial to the success of the agent, then this limitation must be compensated for by the agents networks capacity (i.e., need a bigger brain with memory).
-* Typically the length of the raycasts is limited because the agent need not know about objects that are at the other side of the level. Combined with few raycasts for computational efficiency, this means that an agent may not observe objects that fall between these rays and the issue becomes worse as the objects reduce in size.
+* Typically, the length of the raycasts is limited because the agent need not know about objects that are at the other side of the level. Combined with few raycasts for computational efficiency, this means that an agent may not observe objects that fall between these rays and the issue becomes worse as the objects reduce in size.
-The Camera provides the agent with either a grayscale or an RGB image of the game environment. It goes without saying that there non-linear relationships between nearby pixels in an image. It is this intuition that helps form the basis of Convolutional Neural Networks (CNNs) and established the literature of designing networks that take advantage of these relationships between pixels. Following this established literature of CNNs on image based data, the MLAgent's Camera Sensor provides a means by which the agent can include high dimensional inputs (images) into its observation stream.
+The Camera provides the agent with either a grayscale or an RGB image of the game environment. In many cases, what we want to extract from a set of pixels is invariant to the location of those pixels in the image. It is this intuition that helps form the basis of Convolutional Neural Networks (CNNs) and established the literature of designing networks that take advantage of these relationships between pixels. Following this established literature of CNNs on image based data, the ML-Agent's Camera Sensor provides a means by which the agent can include high dimensional inputs (images) into its observation stream.
-* It requires render the scene and thus is computationally slower than alternatives that do not use rendering
-* It has yet been shown that the Camera Sensor can be used on a headless machine which means it is not yet possible (if at all) to train an agent on a headless infrastructure.
+
+* It requires rendering the scene and thus is computationally slower than alternatives that do not use rendering.
-* The RGB of the camera only provides a maximum of 3 channels to the agent.
+* The RGB of the camera only provides a maximum of three channels to the agent.
-An image can be thought of as a matrix of a predefined width (W) and a height (H) and each pixel can be thought of as simply an array of length 3 (in the case of RGB), `[Red, Green, Blue]` holding the different channel information of the color (channel) intensities at that pixel location. Thus an image is just a 3 dimensional matrix of size WxHx3. A Grid Observation can be thought of as a generalization of this setup where in place of a pixel there is a "cell" which is an array of length N representing different channel intensities at that cell position. From a Convolutional Neural Network point of view, the introduction of multiple channels in an "image" isn't a new concept. One such example is using an RGB-Depth image which is used in several robotics applications. The distinction of Grid Observations is what the data within the channels represents. Instead of limiting the channels to color intensities, the channels within a cell of a Grid Observation generalize to any data that can be represented by a single number (float or int).
-
-Before jumping into the details of the Grid Sensor, an important thing to note is the agent performance and qualitatively different behavior over raycasts. Unity MLAgent's comes with a suite of example environments. One in particular, the [Food Collector](https://github.com/Unity-Technologies/ml-agents/blob/master/docs/Learning-Environment-Examples.md#food-collector), has been the focus of the Grid Sensor development.
+## Overview
-The Food Collector environment can be described as:
-* Set-up: A multi-agent environment where agents compete to collect food.
-* Goal: The agents must learn to collect as many green food spheres as possible while avoiding red spheres.
-* Agents: The environment contains 5 agents with same Behavior Parameters.
+There are three main phases to the observation process of the Grid Sensor:
-When applying the Grid Sensor to this environment, in place of the Raycast Vector Sensor or the Camera Sensor, a Mean Reward of 40-50 is observed. This performance is on par with what is seen by agents trained with RayCasts but the side-by-side comparison of trained agents, shows a qualitative difference in behavior. A deeper study and interpretation of the qualitative differences between agents trained with Raycasts and Vector Sensors verses Grid Sensors is left to future studies.
-
-<img src="images/gridobs-vs-vectorobs.gif" align="middle" width="3000"/>
-
-## Overview
-
-There are 3 main phases to the Grid Sensor:
 1. **Collection** - data is extracted from observed objects
 2. **Encoding** - the extracted data is encoded into a grid observation
 3. **Communication** - the grid observation is sent to python or used by a trained model
 ## Collection
 
-A Grid Sensor is the Grid Observation analog of a Unity Camera but with some notable differences. The sensor is made up of a grid of identical Box Colliders which designate the "cells" of the grid. The Grid Sensor also has a list of "detectable objects" in the form of Unity gameobject tags. When an object that is tagged as a detectable object is present within a cell's Box Collider, that cell is "activated" and a method on the Grid Sensor extracts data from said object and associates that data with the position of the activated cell. Thus the Grid Sensor is always orthographic:
+A Grid Sensor is the Grid Observation analog of a Unity Camera but with some notable differences. The sensor is made up of a grid of identical Box Colliders which designate the "cells" of the grid. The Grid Sensor also has a list of "detectable objects" in the form of Unity GameObject tags. When an object that is tagged as a detectable object is present within a cell's Box Collider, that cell is "activated" and a method on the Grid Sensor extracts data from said object and associates that data with the position of the activated cell. Thus the Grid Sensor is always orthographic:
 
 <img src="images/persp_ortho_proj.png" width="500">
 <cite><a href="https://www.geofx.com/graphics/nehe-three-js/lessons17-24/lesson21/lesson21.html">geofx.com</a></cite>
-Just like the Raycasts mentioned earlier, the Grid Sensor can extract any kind of data from a detected object and just like the Camera, the Grid Sensor maintains the spacial relationship between nearby cells that allows one to take advantage of the CNN literature. Thus the Grid Sensor tries to take the best of both sensors and combines them to something that is more expressive.
+Just like the Raycasts mentioned earlier, the Grid Sensor can extract any kind of data from a detected object, and just like the Camera, the Grid Sensor maintains the spacial relationship between nearby cells that allows one to take advantage of the computational properties of CNNs. Thus the Grid Sensor tries to take the best of both sensors and combines them to something that is more expressive.
-Lets imagine a scenario where an agent is faced with 2 enemies and there are 2 "equipable" weapons somewhat behind the agent. Lets also keep in mind some important properties of the enemies and weapons that would be useful for the agent to know. For simplicity, lets assume enemies represent their health as a percentage (0-100%). Lets also assume that enemies and weapons are the only 2 kind of objects that the agent would see in the entire game.
+Let's imagine a scenario where an agent is faced with two enemies and there are two "equipable" weapons somewhat behind the agent. It would be helpful for the agent to know the location and properties of both the enemies as well as the equippable items. For simplicity, let's assume enemies represent their health as a percentage (0-100%). Also assume that enemies and weapons are the only two kinds of objects that the agent would see in the entire game.
-<img src="images/gridsensor-example.png" align="middle" width="3000"/>
+<img src="images/gridsensor-example.png" align="middle" width="512"/>
-If a raycast hits an object, not only could we get the distance (normalized by the maximum raycast distance) we would be able to extract its type (enemy vs weapon) and if its an enemy then we could get its health (e.g., .6).
+If a raycast hits an object, not only could we get the distance (normalized by the maximum raycast distance) we would be able to extract its type (enemy vs weapon) and any attribute associate with it (e.g. an enemy's health).
 
 There are many ways in which one could encode this information but one reasonable encoding is this:
 ```
 For example, if the raycast hit nothing then this would be represented by `[0, 0, 0, 1]`.
 If instead the raycast hit an enemy with 60% health that is 50% of the maximum raycast distance, the data would be represented by `[0, 1, .6, .5]`.
 
-The limitations of raycasts which were presented above are easy to visualize in the below image. The agent is unable to see where the weapons are and only sees one of the enemies. Typically in the MLAgents examples, this situation is mitigated by including previous frames of data so that the agent observes changes through time. However, in more complex games, it is not difficult to imagine scenarios where an agent would not be able to observe important information using only Raycasts.
+The limitations of raycasts which were presented above are easy to visualize in the below image. The agent is unable to see where the weapons are and only sees one of the enemies. Typically in the ML-Agents examples, this situation is mitigated by including previous frames of data so that the agent observes changes through time. However, in more complex games, it is not difficult to imagine scenarios where an agent might miss important information using only Raycasts.
-<img src="images/gridsensor-example-raycast.png" align="middle" width="3000"/>
+<img src="images/gridsensor-example-raycast.png" align="middle" width="512"/>
-<img src="images/gridsensor-example-camera.png" align="middle" width="3000"/>
+<img src="images/gridsensor-example-camera.png" align="middle" width="512"/>
 
 #### Grid Sensor
 
 
 Following the same data extraction method presented in the section on raycasts, if a Grid Sensor was used instead of Raycasts or a Camera, then not only would the agent be able to extract the health value of the enemies but it would also be able to encode the relative positions of those objects as is done with Camera. Additionally, as the texture of the objects is not used, this data can be collected without rendering the scene.
 
-<img src="images/gridsensor-example-gridsensor.png" align="middle" width="3000"/>
+In our example, we can collect data in the form of [objectType, health] by overriding `GetObjectData` as the following:
+```csharp
+    protected override float[] GetObjectData(GameObject currentColliderGo, float type_index, float normalized_distance)
+    {
+        float[] channelValues = new float[ChannelDepth.Length]; // ChannelDepth.Length = 2 in this example
+        channelValues[0] = type_index; // this is the observation collected in default implementation
+        if (currentColliderGo.tag == "enemy")
+        {
+            var enemy = currentColliderGo.GetComponent<EnemyClass>();
+            channelValues[1] = enemy.health; // the value may have to be normalized depends on the type of GridSensor encoding you use (see sections below)
+        }
+        return channelValues;
+    }
+```
-At the end of the Collection phase, each cell with an object inside of it has `GetObjectData` called and the returned values (named `channelValues`) is then processed in the Encoding phase which is described in the next section.
+<img src="images/gridsensor-example-gridsensor.png" align="middle" width="512"/>
+
+At the end of the Collection phase, each cell with an object inside of it has `GetObjectData` called and the returned values is then processed in the Encoding phase which is described in the next section.
-The CountingGridSensor builds on the GridSesnor to perform the specific job of counting the number of object types that are based on the different detectable object tags. The encoding and is meant to exploit a key feature of the Grid Sensor. In both the Channel and the Channel Hot DepthTypes, the closest detectable object, in relation to the agent, that lays within a cell is used for encoding the value for that cell. In the CountingGridSensor, the number of each type of object is recorded and then normalized according to a max count, stored in the ChannelDepth.
+The CountingGridSensor builds on the GridSensor to perform the specific job of counting the number of object types that are based on the different detectable object tags. The encoding is meant to exploit a key feature of the GridSensor. In original GridSensor, only the closest detectable object, in relation to the agent, that lies within a cell is used for encoding the value for that cell. In the CountingGridSensor, the number of each type of object is recorded and then normalized according to a max count.
 
 An example of the CountingGridSensor can be found below.
 
 In order to support different ways of representing the data extracted from an object, multiple "depth types" were implemented. Each has pros and cons and, depending on the use-case of the Grid Sensor, one may be more beneficial than the others.
 
-The data stored that is extracted during the *Collection* phase, and stored in `channelValues`, may come from different sources. For instance, going back the Enemy/Weapon example in the previous section, an enemy's health is continuous whereas the object type (enemy or weapon) is categorical data. This distinction is important as categorical data requires a different encoding mechanism than continuous data.
+The stored data that is extracted during the *Collection* phase may come from different sources, and thus be of a different nature. For instance, going back to the Enemy/Weapon example in the previous section, an enemy's health is continuous whereas the object type (enemy or weapon) is categorical data. This distinction is important as categorical data requires a different encoding mechanism than continuous data.
-The Grid Sensor handles this distinction with 4 properties that define how this data is to be encoded:
-* DepthType - Enum signifying the encoding mode: Channel, ChannelHot
-* ObservationPerCell - the total number of values that are in each cell of the grid observation
-* ChannelDepth - int[] describing the range of each data within the `channelValues`
-* ChannelOffset - int[] describing the number of encoded values that come before each data within `channelValues`
+The GridSensor handles this distinction with two user defined properties that define how this data is to be encoded:
-The ChannelDepth and the DepthType are user defined and gives control to the developer to how they can encode their data. The ChannelDepth and ChannelOffset are both initialized and used in different ways depending on the ChannelDepth and the DepthType.
+* DepthType - Enum signifying the encoding mode: Channel, ChannelHot
+* ChannelDepth - `int[]` describing the range of each data and is used differently with different DepthType
-How categorical and continuous data is treated is different between the different DepthTypes as will be explored in the sections below. The sections will use an on-going example similar to example mentioned earlier where, within a cell, the sensor observes: `an enemy with 60% health`. Thus the cell contains 2 kinds of data: categorical data (object type) and the continuous data (health). Additionally, the order of the observed tags is important as it allows one to encode the tag of the observed object by its index within list of observed tags. Note that in the example, the observed tags is defined as ["weapon", "enemy"].
+How categorical and continuous data is treated is different between the different DepthTypes as will be explored in the sections below. The sections will use an on-going example similar to the example mentioned earlier where, within a cell, the sensor observes: `an enemy with 60% health`. Thus the cell contains two kinds of data: categorical data (object type) and the continuous data (health). Additionally, the order of the observed tags is important as it allows one to encode the tag of the observed object by its index within the list of observed tags. Note that in the example, the observed tags is defined as ["weapon", "enemy"].
-The Channel Based Grid Observations is perhaps the simplest in terms of usability and similarity with other machine learning applications. Each grid is of size WxHxC where C is the number of channels. To distinguish between categorical and continuous data, one would use the ChannelDepth array to signify the ranges that the values in the `channelValues` array could take. If one sets ChannelDepth[i] to be 1, it is assumed that the value of `channelValues[i]` is already normalized. Else ChannelDepth[i] represents the total number of possible values that `channelValues[i]` can take.
+The Channel Based Grid Observations represent obsevations in a normalized form with 0 to 1. To distinguish between categorical and continuous data, one would use the ChannelDepth array to signify the ranges that the values in the `channelValues` array could take. If one sets ChannelDepth[i] to be 1, it is assumed that the value of `channelValues[i]` is already normalized. Else ChannelDepth[i] represents the total number of possible values that `channelValues[i]` can take and will be used for normalization.
-Using the example described earlier, if one was using Channel Based Grid Observations, they would have a ChannelDepth = {2, 1} to describe that there are two possible values for the first channel and the 1 represents that the second channel is already normalized.
-As the "enemy" is in the second position of the observed tags, its value can be normalized by:
+For continuous data, you should specify `ChannelDepth[i]` to 1 and the collected data should be already normalized by its min/max range. For discrete data, you should specify `ChannelDepth[i]` to be the total number of possible values, and the collected data should be an integer value within range of `ChannelDepth[i]`.
+
+Using the example described earlier, if one was using Channel Based Grid Observations, they would have a ChannelDepth = {2, 1} to describe that there are two possible values for the first channel (ObjectType) and the 1 represents that the second channel (EnemyHealth) is continuous and should be already normalized.
+
+For ObjectType, "weapon", "enemy" will be represented respectively as:
-num = detectableObjects.IndexOfTag("enemy")/ChannelDepth[0] = 2/2 = 1;
+weapon = DetectableObjects.IndexOfTag("weapon")/ChannelDepth[0] = 1/2 = 0.5;
+enemy = DetectableObjects.IndexOfTag("enemy")/ChannelDepth[0] = 2/2 = 1;
-`[1, .6]`
+`[1, .6]`. If the health in the game is not represented in a normalized form, for example if the health is represented in an integer ranging from -100 to 100, you'll need to manully nomalize it during collection. That is, If you get value 50, you need to normalize it by `50/(100- (-100))=0.25` and collect 0.25 instead of 50.
-The Channel Hot DepthType generalizes the classic OneHot encoding to differentiate combinations of different data. Rather than normalizing the data like in the Channel Based section, each element of `channelValues` is represented by an encoding based on the ChannelDepth. If ChannelDepth[i] = 1, then this represents that `channelValues[i]` is already normalized (between 0-1) and will be used directly within the encoding. However if ChannelDepth[i] is an integer greater than 1, then the value in `channelValues[i]` will be converted into a OneHot encoding based on the following:
+The Channel Hot DepthType generalizes the classic OneHot encoding to differentiate combinations of different data. Rather than normalizing the data like in the Channel Based section, each element of `channelValues` is represented by an encoding based on the ChannelDepth. If ChannelDepth[i] = 1, then this represents that `channelValues[i]` is already normalized (between 0-1) and will be used directly within the encoding which is same as with Channel Based. However if ChannelDepth[i] is an integer greater than 1, then the value in `channelValues[i]` will be converted into a OneHot encoding based on the following:
 
 ```
 float[] arr = new float[ChannelDepth[i] + 1];
 The encoding of each channel is then concatenated together. Clearly using this setup allows the developer to be able to encode values using the classic OneHot encoding. Below are some different variations of the ChannelDepth which create different encodings of the example:
 
 ##### ChannelDepth = {3, 1}
-The first element, 3, signifies that there are 3 possibilities for the first channel and as the "enemy" is 2nd in the detected objects list, the "enemy" in the example is encoded as `[0, 0, 1]` where the first index represents "no object". The second element, 1, signifies that the health is already normalized and, following the table, is used directly. The resulting encoding is thus:
+The first element, 3, signifies that there are three possibilities for the first channel and as the "enemy" is 2nd in the detected objects list, the "enemy" in the example is encoded as `[0, 0, 1]` where the first index represents "no object". The second element, 1, signifies that the health is already normalized and, following the table, is used directly. The resulting encoding is thus:
 ```
 [0, 0, 1, 0.6]
 ```
 
 ### CountingGridSensor
 
-As introduced above, the CountingGridSensor inherits from the GridSensor for the sole purpose of counting the different objects that lay within a cell. In order to normalize the counts so that the grid can be properly encoded as PNG, the ChannelDepth is used to represent the "maximum count" of each type. For the working example, if the ChannelDepth is set as {50, 10}, which represents that the maximum count for objects with the "weapon" and "enemy" tag is 50 and 10, respectively, then the resulting data would be:
+As mentioned above, the CountingGridSensor inherits from the GridSensor for the sole purpose of counting the different objects that lay within a cell. In order to normalize the counts so that the grid can be properly encoded as PNG, the ChannelDepth is used to represent the "maximum count" of each type. For the working example, if the ChannelDepth is set as {50, 10}, which represents that the maximum count for objects with the "weapon" and "enemy" tag is 50 and 10, respectively, then the resulting data would be:
 ```
 encoding = [0 weapons/ 50 weapons, 1 enemy / 10 enemies] = [0, .1]
 ```
-At the end of the Encoding phase, all of the data for a Grid Observation is placed into a float[] referred to as the perception buffer. Now the data is ready to be sent to either the python side for training or to be used by a trained model within Unity. This is where the Grid Sensor takes advantage of 2D textures and the PNG encoding schema to reduce the number of bytes that are being sent.
+At the end of the Encoding phase, all the Grid Observations will be sent to either the python side for training or to be used by a trained model within Unity. Since the data format is similar to images collected by Camera Sensors, Grid Observations also have the CompressionType option to specify whether to send the data directly or send in PNG compressed form for better communication efficiency.
-The 2D texture is a Unity class that encodes the colors of an image. It is used for many ways through out Unity but it has 2 specific methods that the Grid Sensor takes advantage of:
-
-`SetPixels` takes a 2D array of Colors and assigns the color values to the texture.
-
-`EncodeToPNG` returns a byte[] containing the PNG encoding of the colors of the texture.
-
-Together these 2 functions allow one to "push" a WxHx3 normalized array to a PNG byte[]. And indeed, this is how the Camera Sensor in Unity MLAgents sends its data to python. However, the grid sensor can have N channels so there needs to be a more generic way to send the data.
-
-The core idea behind how a Grid Observation is encoded is the following:
-1. split the channels of a Grid Observation into groups of 3
-2. encode each of these groups as a PNG byte[]
-3. concatenate all byte[] and send the combined array to python
-4. reconstruct the Grid Observation by splitting up the array and decoding the sections
-
-Once the bytes are sent to python, they are then decoded and used as a tensor of the correct shape within the mlagents python codebase.
+Once the bytes are sent to Python, they are then decoded and provided as a tensor of the correct shape.

com.unity.ml-agents.extensions/Documentation~/com.unity.ml-agents.extensions.md

 | _Runtime_        | Contains core C# APIs for integrating ML-Agents into your Unity scene. |
 | _Tests_          | Contains the unit tests for the package.                               |
 
+The Runtime directory currently contains three features:
+ * [Match-3 sensor and actuator](Match3.md)
+ * [Grid-based sensor](Grid-Sensor.md)
+ * Physics-based sensors
+ * [Input System Package Integration](InputActuatorComponent.md)
+
-[Clone the repository](../../docs/Installation.md#clone-the-ml-agents-toolkit-repository-optional) and follow the
-[Local Installation for Development](../../docs/Installation.md#advanced-local-installation-for-development-1)
+[Clone the repository](https://github.com/Unity-Technologies/ml-agents/tree/release_12_docs/docs/Installation.md#clone-the-ml-agents-toolkit-repository-optional) and follow the
+[Local Installation for Development](https://github.com/Unity-Technologies/ml-agents/tree/release_12_docs/docs/Installation.md#advanced-local-installation-for-development-1)
-![Package Manager git URL](../../docs/images/unity_package_manager_git_url.png)
+![Package Manager git URL](https://github.com/Unity-Technologies/ml-agents/blob/release_12_docs/docs/images/unity_package_manager_git_url.png)
+
 In the dialog that appears, enter
  ```
  git+https://github.com/Unity-Technologies/ml-agents.git?path=com.unity.ml-agents.extensions
 This version of the Unity ML-Agents Extensions package is compatible with the
 following versions of the Unity Editor:
 
-- 2018.4 and later
+- If using the `InputActuatorComponent`
+    - 2019.4 or later
+    - install the `com.unity.inputsystem` package version `1.1.0-preview.3` or later.
+- Else 2018.4 and later
-none
+- For the `InputActuatorComponent`
+    - Limited implementation of `InputControls`
+    - No way to customize the action space of the `InputActuatorComponent`
-The main [README](../../README.md) contains links for contacting the team or getting support.
+The main [README](https://github.com/Unity-Technologies/ml-agents/tree/release_12_docs/README.md) contains links for contacting the team or getting support.

com.unity.ml-agents.extensions/Editor/EditorExample.cs

-using Unity.MLAgents;
-
 namespace Unity.MLAgents.Extensions.Editor
 {
 

com.unity.ml-agents.extensions/Editor/RigidBodySensorComponentEditor.cs

-using UnityEngine;
 using UnityEditor;
 using Unity.MLAgents.Editor;
 using Unity.MLAgents.Extensions.Sensors;

com.unity.ml-agents.extensions/Runtime/Sensors/CountingGridSensor.cs

 using System;
 using UnityEngine;
-using UnityEngine.Assertions;
 
 namespace Unity.MLAgents.Extensions.Sensors
 {

com.unity.ml-agents.extensions/Runtime/Sensors/GridSensor.cs

     /// <summary>
     /// Grid-based sensor.
     /// </summary>
-    public class GridSensor : SensorComponent, ISensor
+    public class GridSensor : SensorComponent, ISensor, IBuiltInSensor
     {
         /// <summary>
         /// Name of this grid sensor.
         {
             return CompressionType;
         }
+
+        /// <inheritdoc/>
+        public BuiltInSensorType GetBuiltInSensorType()
+        {
+            return BuiltInSensorType.GridSensor;
+        }
+
 
         /// <summary>
         /// GetCompressedObservation - Calls Perceive then puts the data stored on the perception buffer

com.unity.ml-agents.extensions/Runtime/Sensors/PhysicsBodySensor.cs

 using System.Collections.Generic;
+#if UNITY_2020_1_OR_NEWER
+#endif
 using Unity.MLAgents.Sensors;
 
 namespace Unity.MLAgents.Extensions.Sensors
     /// </summary>
-    public class PhysicsBodySensor : ISensor
+    public class PhysicsBodySensor : ISensor, IBuiltInSensor
     {
         int[] m_Shape;
         string m_SensorName;
         {
             return m_SensorName;
         }
+
+        /// <inheritdoc/>
+        public BuiltInSensorType GetBuiltInSensorType()
+        {
+            return BuiltInSensorType.PhysicsBodySensor;
+        }
+
     }
 }

com.unity.ml-agents.extensions/Runtime/Sensors/RigidBodyJointExtractor.cs

-using System.Collections.Generic;
 using UnityEngine;
 using Unity.MLAgents.Sensors;
 

com.unity.ml-agents.extensions/Runtime/Unity.ML-Agents.Extensions.asmdef

     "name": "Unity.ML-Agents.Extensions",
     "references": [
         "Unity.Barracuda",
-        "Unity.ML-Agents"
+        "Unity.ML-Agents",
+        "Unity.ML-Agents.Extensions.Input"
     ],
     "includePlatforms": [],
     "excludePlatforms": []

com.unity.ml-agents.extensions/Tests/Editor/Sensors/ChannelHotPerceiveTests.cs

         {
             string[] tags = { k_Tag1 };
             int[] depths = { 1 };
-            dummyData.Data = new float[] { -0.1f };
+            dummyData.Data = new[] { -0.1f };
             Color[] colors = { Color.red, Color.magenta };
             gridSensor.SetParameters(tags, depths, GridSensor.GridDepthType.ChannelHot,
                 1f, 1f, 10, 10, LayerMask.GetMask("Default"), false, colors);
         {
             string[] tags = { k_Tag1 };
             int[] depths = { 1 };
-            dummyData.Data = new float[] { 1.1f };
+            dummyData.Data = new[] { 1.1f };
             Color[] colors = { Color.red, Color.magenta };
             gridSensor.SetParameters(tags, depths, GridSensor.GridDepthType.ChannelHot,
                 1f, 1f, 10, 10, LayerMask.GetMask("Default"), false, colors);
         {
             string[] tags = { k_Tag1 };
             int[] depths = { 1 };
-            dummyData.Data = new float[] { .2f };
+            dummyData.Data = new[] { .2f };
             Color[] colors = { Color.red, Color.magenta };
             gridSensor.SetParameters(tags, depths, GridSensor.GridDepthType.ChannelHot,
                 1f, 1f, 10, 10, LayerMask.GetMask("Default"), false, colors);
             Assert.AreEqual(10 * 10 * 1, output.Length);
 
             int[] subarrayIndicies = new int[] { 77, 78, 87, 88 };
-            float[][] expectedSubarrays = GridObsTestUtils.DuplicateArray(new float[] { .2f }, 4);
-            float[] expectedDefault = new float[] { 0 };
+            float[][] expectedSubarrays = GridObsTestUtils.DuplicateArray(new[] { .2f }, 4);
+            float[] expectedDefault = new[] { 0.0f };
             GridObsTestUtils.AssertSubarraysAtIndex(output, subarrayIndicies, expectedSubarrays, expectedDefault);
         }
 
             string[] tags = { k_Tag1 };
             int[] depths = { 3 };
-            dummyData.Data = new float[] { -1f };
+            dummyData.Data = new[] { -1f };
             Color[] colors = { Color.red, Color.magenta };
             gridSensor.SetParameters(tags, depths, GridSensor.GridDepthType.ChannelHot,
                 1f, 1f, 10, 10, LayerMask.GetMask("Default"), false, colors);
         {
             string[] tags = { k_Tag1 };
             int[] depths = { 3 };
-            dummyData.Data = new float[] { 4f };
+            dummyData.Data = new[] { 4f };
             Color[] colors = { Color.red, Color.magenta };
             gridSensor.SetParameters(tags, depths, GridSensor.GridDepthType.ChannelHot,
                 1f, 1f, 10, 10, LayerMask.GetMask("Default"), false, colors);
         {
             string[] tags = { k_Tag1 };
             int[] depths = { 3 };
-            dummyData.Data = new float[] { 2f };
+            dummyData.Data = new[] { 2f };
             Color[] colors = { Color.red, Color.magenta };
             gridSensor.SetParameters(tags, depths, GridSensor.GridDepthType.ChannelHot,
                 1f, 1f, 10, 10, LayerMask.GetMask("Default"), false, colors);
 
             Assert.AreEqual(10 * 10 * 3, output.Length);
 
-            int[] subarrayIndicies = new int[] { 77, 78, 87, 88 };
+            int[] subarrayIndicies = new[] { 77, 78, 87, 88 };
-            float[] expectedDefault = new float[] { 1, 0, 0 };
+            float[] expectedDefault = new[] { 1.0f, 0.0f, 0.0f };
             GridObsTestUtils.AssertSubarraysAtIndex(output, subarrayIndicies, expectedSubarrays, expectedDefault);
         }
 
             string[] tags = { k_Tag1 };
             int[] depths = { 3 };
-            dummyData.Data = new float[] { 2.4f };
+            dummyData.Data = new[] { 2.4f };
             Color[] colors = { Color.red, Color.magenta };
             gridSensor.SetParameters(tags, depths, GridSensor.GridDepthType.ChannelHot,
                 1f, 1f, 10, 10, LayerMask.GetMask("Default"), false, colors);
         {
             string[] tags = { k_Tag1 };
             int[] depths = { 1, 1 };
-            dummyData.Data = new float[] { .4f, .3f };
+            dummyData.Data = new[] { .4f, .3f };
             Color[] colors = { Color.red, Color.magenta };
             gridSensor.SetParameters(tags, depths, GridSensor.GridDepthType.ChannelHot,
                 1f, 1f, 10, 10, LayerMask.GetMask("Default"), false, colors);
             Assert.AreEqual(10 * 10 * 2, output.Length);
 
             int[] subarrayIndicies = new int[] { 77, 78, 87, 88 };
-            float[][] expectedSubarrays = GridObsTestUtils.DuplicateArray(new float[] { .4f, .3f }, 4);
+            float[][] expectedSubarrays = GridObsTestUtils.DuplicateArray(new[] { .4f, .3f }, 4);
             float[] expectedDefault = new float[] { 0, 0 };
             GridObsTestUtils.AssertSubarraysAtIndex(output, subarrayIndicies, expectedSubarrays, expectedDefault);
         }
         {
             string[] tags = { k_Tag1 };
             int[] depths = { 1, 3 };
-            dummyData.Data = new float[] { .4f, 4f };
+            dummyData.Data = new[] { .4f, 4f };
             Color[] colors = { Color.red, Color.magenta };
             gridSensor.SetParameters(tags, depths, GridSensor.GridDepthType.ChannelHot,
                 1f, 1f, 10, 10, LayerMask.GetMask("Default"), false, colors);
         {
             string[] tags = { k_Tag1 };
             int[] depths = { 1, 3 };
-            dummyData.Data = new float[] { .4f, 1f };
+            dummyData.Data = new[] { .4f, 1f };
             Color[] colors = { Color.red, Color.magenta };
             gridSensor.SetParameters(tags, depths, GridSensor.GridDepthType.ChannelHot,
                 1f, 1f, 10, 10, LayerMask.GetMask("Default"), false, colors);
             Assert.AreEqual(10 * 10 * 4, output.Length);
 
             int[] subarrayIndicies = new int[] { 77, 78, 87, 88 };
-            float[][] expectedSubarrays = GridObsTestUtils.DuplicateArray(new float[] { .4f, 0, 1, 0 }, 4);
+            float[][] expectedSubarrays = GridObsTestUtils.DuplicateArray(new[] { .4f, 0, 1, 0 }, 4);
             float[] expectedDefault = new float[] { 0, 1, 0, 0 };
             GridObsTestUtils.AssertSubarraysAtIndex(output, subarrayIndicies, expectedSubarrays, expectedDefault);
         }
         {
             string[] tags = { k_Tag1 };
             int[] depths = { 3, 1 };
-            dummyData.Data = new float[] { 1f, .4f };
+            dummyData.Data = new[] { 1f, .4f };
             Color[] colors = { Color.red, Color.magenta };
             gridSensor.SetParameters(tags, depths, GridSensor.GridDepthType.ChannelHot,
                 1f, 1f, 10, 10, LayerMask.GetMask("Default"), false, colors);
 
             Assert.AreEqual(10 * 10 * 4, output.Length);
 
-            int[] subarrayIndicies = new int[] { 77, 78, 87, 88 };
-            float[][] expectedSubarrays = GridObsTestUtils.DuplicateArray(new float[] { 0, 1, 0, .4f }, 4);
+            int[] subarrayIndicies = new[] { 77, 78, 87, 88 };
+            float[][] expectedSubarrays = GridObsTestUtils.DuplicateArray(new[] { 0, 1, 0, .4f }, 4);
             float[] expectedDefault = new float[] { 1, 0, 0, 0 };
             GridObsTestUtils.AssertSubarraysAtIndex(output, subarrayIndicies, expectedSubarrays, expectedDefault);
         }
         {
             string[] tags = { k_Tag1 };
             int[] depths = { 3, 3 };
-            dummyData.Data = new float[] { 1f, 2.2f };
+            dummyData.Data = new[] { 1f, 2.2f };
             Color[] colors = { Color.red, Color.magenta };
             gridSensor.SetParameters(tags, depths, GridSensor.GridDepthType.ChannelHot,
                 1f, 1f, 10, 10, LayerMask.GetMask("Default"), false, colors);
         {
             string[] tags = { k_Tag1 };
             int[] depths = { 5, 1, 3 };
-            dummyData.Data = new float[] { 3f, .6f, 2.2f };
+            dummyData.Data = new[] { 3f, .6f, 2.2f };
             Color[] colors = { Color.red, Color.magenta };
             gridSensor.SetParameters(tags, depths, GridSensor.GridDepthType.ChannelHot,
                 1f, 1f, 10, 10, LayerMask.GetMask("Default"), false, colors);
             Assert.AreEqual(10 * 10 * 9, output.Length);
 
             int[] subarrayIndicies = new int[] { 77, 78, 87, 88 };
-            float[][] expectedSubarrays = GridObsTestUtils.DuplicateArray(new float[] { 0, 0, 0, 1, 0, .6f, 0, 0, 1 }, 4);
+            float[][] expectedSubarrays = GridObsTestUtils.DuplicateArray(new[] { 0, 0, 0, 1, 0, .6f, 0, 0, 1 }, 4);
             float[] expectedDefault = new float[] { 1, 0, 0, 0, 0, 0, 1, 0, 0 };
             GridObsTestUtils.AssertSubarraysAtIndex(output, subarrayIndicies, expectedSubarrays, expectedDefault);
         }
         {
             string[] tags = { k_Tag1 };
             int[] depths = { 5, 1, 3 };
-            dummyData.Data = new float[] { 3f, .6f, 2.2f };
+            dummyData.Data = new[] { 3f, .6f, 2.2f };
             Color[] colors = { Color.red, Color.magenta };
             gridSensor.SetParameters(tags, depths, GridSensor.GridDepthType.ChannelHot,
                 1f, 1f, 10, 10, LayerMask.GetMask("Default"), false, colors);
             Assert.AreEqual(10 * 10 * 9, output.Length);
 
             int[] subarrayIndicies = new int[] { 77, 78, 87, 88 };
-            float[][] expectedSubarrays = GridObsTestUtils.DuplicateArray(new float[] { 0, 0, 0, 1, 0, .6f, 0, 0, 1 }, 4);
+            float[][] expectedSubarrays = GridObsTestUtils.DuplicateArray(new[] { 0, 0, 0, 1, 0, .6f, 0, 0, 1 }, 4);
             float[] expectedDefault = new float[] { 1, 0, 0, 0, 0, 0, 1, 0, 0 };
             GridObsTestUtils.AssertSubarraysAtIndex(output, subarrayIndicies, expectedSubarrays, expectedDefault);
 

com.unity.ml-agents.extensions/Tests/Editor/Sensors/ChannelHotShapeTests.cs

 using NUnit.Framework;
 using UnityEngine;
 using Unity.MLAgents.Extensions.Sensors;
-using NUnit.Framework.Internal;
-using UnityEngine.TestTools;
-using System.Collections;
 
 namespace Unity.MLAgents.Extensions.Tests.Sensors
 {

com.unity.ml-agents.extensions/Tests/Editor/Sensors/ChannelPerceiveTests.cs

 using UnityEngine.TestTools;
 using Unity.MLAgents.Extensions.Sensors;
 using Unity.MLAgents.Extensions.TestUtils.Sensors;
-using System.Linq;
 
 namespace Unity.MLAgents.Extensions.Tests.Sensors
 {
         {
             string[] tags = { k_Tag1 };
             int[] depths = { 1 };
-            dummyData.Data = new float[] { -0.1f };
+            dummyData.Data = new[] { -0.1f };
             Color[] colors = { Color.red, Color.magenta };
             gridSensor.SetParameters(tags, depths, GridSensor.GridDepthType.Channel,
                 1f, 1f, 10, 10, LayerMask.GetMask("Default"), false, colors);
         {
             string[] tags = { k_Tag1 };
             int[] depths = { 1 };
-            dummyData.Data = new float[] { 1.1f };
+            dummyData.Data = new[] { 1.1f };
             Color[] colors = { Color.red, Color.magenta };
             gridSensor.SetParameters(tags, depths, GridSensor.GridDepthType.Channel,
                 1f, 1f, 10, 10, LayerMask.GetMask("Default"), false, colors);
         {
             string[] tags = { k_Tag1 };
             int[] depths = { 1 };
-            dummyData.Data = new float[] { .2f };
+            dummyData.Data = new[] { .2f };
             Color[] colors = { Color.red, Color.magenta };
             gridSensor.SetParameters(tags, depths, GridSensor.GridDepthType.Channel,
                 1f, 1f, 10, 10, LayerMask.GetMask("Default"), false, colors);
             Assert.AreEqual(10 * 10 * 1, output.Length);
 
             int[] subarrayIndicies = new int[] { 77, 78, 87, 88 };
-            float[][] expectedSubarrays = GridObsTestUtils.DuplicateArray(new float[] { .2f }, 4);
+            float[][] expectedSubarrays = GridObsTestUtils.DuplicateArray(new[] { .2f }, 4);
             float[] expectedDefault = new float[] { 0 };
             GridObsTestUtils.AssertSubarraysAtIndex(output, subarrayIndicies, expectedSubarrays, expectedDefault);
         }
         {
             string[] tags = { k_Tag1 };
             int[] depths = { 3 };
-            dummyData.Data = new float[] { -1f };
+            dummyData.Data = new[] { -1f };
             Color[] colors = { Color.red, Color.magenta };
             gridSensor.SetParameters(tags, depths, GridSensor.GridDepthType.Channel,
                 1f, 1f, 10, 10, LayerMask.GetMask("Default"), false, colors);
         {
             string[] tags = { k_Tag1 };
             int[] depths = { 3 };
-            dummyData.Data = new float[] { 4f };
+            dummyData.Data = new[] { 4f };
             Color[] colors = { Color.red, Color.magenta };
             gridSensor.SetParameters(tags, depths, GridSensor.GridDepthType.Channel,
                 1f, 1f, 10, 10, LayerMask.GetMask("Default"), false, colors);
         {
             string[] tags = { k_Tag1 };
             int[] depths = { 3 };
-            dummyData.Data = new float[] { 2f };
+            dummyData.Data = new[] { 2f };
             Color[] colors = { Color.red, Color.magenta };
             gridSensor.SetParameters(tags, depths, GridSensor.GridDepthType.Channel,
                 1f, 1f, 10, 10, LayerMask.GetMask("Default"), false, colors);
             Assert.AreEqual(10 * 10 * 1, output.Length);
 
             int[] subarrayIndicies = new int[] { 77, 78, 87, 88 };
-            float[][] expectedSubarrays = GridObsTestUtils.DuplicateArray(new float[] { 2f / 3f }, 4);
+            float[][] expectedSubarrays = GridObsTestUtils.DuplicateArray(new[] { 2f / 3f }, 4);
             float[] expectedDefault = new float[] { 0 };
             GridObsTestUtils.AssertSubarraysAtIndex(output, subarrayIndicies, expectedSubarrays, expectedDefault);
         }
         {
             string[] tags = { k_Tag1 };
             int[] depths = { 3 };
-            dummyData.Data = new float[] { 2.4f };
+            dummyData.Data = new[] { 2.4f };
             Color[] colors = { Color.red, Color.magenta };
             gridSensor.SetParameters(tags, depths, GridSensor.GridDepthType.Channel,
                 1f, 1f, 10, 10, LayerMask.GetMask("Default"), false, colors);
             Assert.AreEqual(10 * 10 * 1, output.Length);
 
             int[] subarrayIndicies = new int[] { 77, 78, 87, 88 };
-            float[][] expectedSubarrays = GridObsTestUtils.DuplicateArray(new float[] { 2.4f / 3f }, 4);
+            float[][] expectedSubarrays = GridObsTestUtils.DuplicateArray(new[] { 2.4f / 3f }, 4);
             float[] expectedDefault = new float[] { 0 };
             GridObsTestUtils.AssertSubarraysAtIndex(output, subarrayIndicies, expectedSubarrays, expectedDefault);
         }
         {
             string[] tags = { k_Tag1 };
             int[] depths = { 1, 1 };
-            dummyData.Data = new float[] { .4f, .3f };
+            dummyData.Data = new[] { .4f, .3f };
             Color[] colors = { Color.red, Color.magenta };
             gridSensor.SetParameters(tags, depths, GridSensor.GridDepthType.Channel,
                 1f, 1f, 10, 10, LayerMask.GetMask("Default"), false, colors);
             Assert.AreEqual(10 * 10 * 2, output.Length);
 
             int[] subarrayIndicies = new int[] { 77, 78, 87, 88 };
-            float[][] expectedSubarrays = GridObsTestUtils.DuplicateArray(new float[] { .4f, .3f }, 4);
+            float[][] expectedSubarrays = GridObsTestUtils.DuplicateArray(new[] { .4f, .3f }, 4);
             float[] expectedDefault = new float[] { 0, 0 };
             GridObsTestUtils.AssertSubarraysAtIndex(output, subarrayIndicies, expectedSubarrays, expectedDefault);
         }
         {
             string[] tags = { k_Tag1 };
             int[] depths = { 1, 3 };
-            dummyData.Data = new float[] { .4f, 4f };
+            dummyData.Data = new[] { .4f, 4f };
             Color[] colors = { Color.red, Color.magenta };
             gridSensor.SetParameters(tags, depths, GridSensor.GridDepthType.Channel,
                 1f, 1f, 10, 10, LayerMask.GetMask("Default"), false, colors);
         {
             string[] tags = { k_Tag1 };
             int[] depths = { 1, 3 };
-            dummyData.Data = new float[] { .4f, 1f };
+            dummyData.Data = new[] { .4f, 1f };
             Color[] colors = { Color.red, Color.magenta };
             gridSensor.SetParameters(tags, depths, GridSensor.GridDepthType.Channel,
                 1f, 1f, 10, 10, LayerMask.GetMask("Default"), false, colors);
             Assert.AreEqual(10 * 10 * 2, output.Length);
 
             int[] subarrayIndicies = new int[] { 77, 78, 87, 88 };
-            float[][] expectedSubarrays = GridObsTestUtils.DuplicateArray(new float[] { .4f, 1f / 3f }, 4);
+            float[][] expectedSubarrays = GridObsTestUtils.DuplicateArray(new[] { .4f, 1f / 3f }, 4);
             float[] expectedDefault = new float[] { 0, 0 };
             GridObsTestUtils.AssertSubarraysAtIndex(output, subarrayIndicies, expectedSubarrays, expectedDefault);
         }
         {
             string[] tags = { k_Tag1 };
             int[] depths = { 3, 1 };
-            dummyData.Data = new float[] { 1f, .4f };
+            dummyData.Data = new[] { 1f, .4f };
             Color[] colors = { Color.red, Color.magenta };
             gridSensor.SetParameters(tags, depths, GridSensor.GridDepthType.Channel,
                 1f, 1f, 10, 10, LayerMask.GetMask("Default"), false, colors);
             Assert.AreEqual(10 * 10 * 2, output.Length);
 
             int[] subarrayIndicies = new int[] { 77, 78, 87, 88 };
-            float[][] expectedSubarrays = GridObsTestUtils.DuplicateArray(new float[] { 1f / 3f, .4f }, 4);
+            float[][] expectedSubarrays = GridObsTestUtils.DuplicateArray(new[] { 1f / 3f, .4f }, 4);
             float[] expectedDefault = new float[] { 0, 0 };
             GridObsTestUtils.AssertSubarraysAtIndex(output, subarrayIndicies, expectedSubarrays, expectedDefault);
         }
         {
             string[] tags = { k_Tag1 };
             int[] depths = { 3, 3 };
-            dummyData.Data = new float[] { 1f, 2.2f };
+            dummyData.Data = new[] { 1f, 2.2f };
             Color[] colors = { Color.red, Color.magenta };
             gridSensor.SetParameters(tags, depths, GridSensor.GridDepthType.Channel,
                 1f, 1f, 10, 10, LayerMask.GetMask("Default"), false, colors);
             Assert.AreEqual(10 * 10 * 2, output.Length);
 
             int[] subarrayIndicies = new int[] { 77, 78, 87, 88 };
-            float[][] expectedSubarrays = GridObsTestUtils.DuplicateArray(new float[] { 1f / 3f, 2.2f / 3 }, 4);
+            float[][] expectedSubarrays = GridObsTestUtils.DuplicateArray(new[] { 1f / 3f, 2.2f / 3 }, 4);
             float[] expectedDefault = new float[] { 0, 0 };
             GridObsTestUtils.AssertSubarraysAtIndex(output, subarrayIndicies, expectedSubarrays, expectedDefault);
         }
         {
             string[] tags = { k_Tag1 };
             int[] depths = { 5, 1, 3 };
-            dummyData.Data = new float[] { 3f, .6f, 2.2f };
+            dummyData.Data = new[] { 3f, .6f, 2.2f };
             Color[] colors = { Color.red, Color.magenta };
             gridSensor.SetParameters(tags, depths, GridSensor.GridDepthType.Channel,
                 1f, 1f, 10, 10, LayerMask.GetMask("Default"), false, colors);
             Assert.AreEqual(10 * 10 * 3, output.Length);
 
             int[] subarrayIndicies = new int[] { 77, 78, 87, 88 };
-            float[][] expectedSubarrays = GridObsTestUtils.DuplicateArray(new float[] { 3f / 5f, .6f, 2.2f / 3f }, 4);
+            float[][] expectedSubarrays = GridObsTestUtils.DuplicateArray(new[] { 3f / 5f, .6f, 2.2f / 3f }, 4);
             float[] expectedDefault = new float[] { 0, 0, 0 };
             GridObsTestUtils.AssertSubarraysAtIndex(output, subarrayIndicies, expectedSubarrays, expectedDefault);
         }
         {
             string[] tags = { k_Tag1 };
             int[] depths = { 5, 1, 3 };
-            dummyData.Data = new float[] { 3f, .6f, 2.2f };
+            dummyData.Data = new[] { 3f, .6f, 2.2f };
             Color[] colors = { Color.red, Color.magenta };
             gridSensor.SetParameters(tags, depths, GridSensor.GridDepthType.Channel,
                 1f, 1f, 10, 10, LayerMask.GetMask("Default"), false, colors);
             Assert.AreEqual(10 * 10 * 3, output.Length);
 
             int[] subarrayIndicies = new int[] { 77, 78, 87, 88 };
-            float[][] expectedSubarrays = GridObsTestUtils.DuplicateArray(new float[] { 3f / 5f, .6f, 2.2f / 3f }, 4);
+            float[][] expectedSubarrays = GridObsTestUtils.DuplicateArray(new[] { 3f / 5f, .6f, 2.2f / 3f }, 4);
             float[] expectedDefault = new float[] { 0, 0, 0 };
             GridObsTestUtils.AssertSubarraysAtIndex(output, subarrayIndicies, expectedSubarrays, expectedDefault);
 

com.unity.ml-agents.extensions/Tests/Editor/Sensors/ChannelShapeTests.cs

 using NUnit.Framework;
 using UnityEngine;
 using Unity.MLAgents.Extensions.Sensors;
-using NUnit.Framework.Internal;
 
 namespace Unity.MLAgents.Extensions.Tests.Sensors
 {

com.unity.ml-agents.extensions/Tests/Editor/Sensors/CountingGridSensorPerceiveTests.cs

             Assert.AreEqual(10 * 10 * 1, output.Length);
 
             int[] subarrayIndicies = new int[] { 77, 78, 87, 88 };
-            float[][] expectedSubarrays = GridObsTestUtils.DuplicateArray(new float[] { 1f }, 4);
+            float[][] expectedSubarrays = GridObsTestUtils.DuplicateArray(new[] { 1f }, 4);
             float[] expectedDefault = new float[] { 0 };
             GridObsTestUtils.AssertSubarraysAtIndex(output, subarrayIndicies, expectedSubarrays, expectedDefault);
         }
             Assert.AreEqual(10 * 10 * 1, output.Length);
 
             int[] subarrayIndicies = new int[] { 77, 78, 87, 88 };
-            float[][] expectedSubarrays = GridObsTestUtils.DuplicateArray(new float[] { 1f }, 4);
-            float[] expectedDefault = new float[] { 0 };
+            float[][] expectedSubarrays = GridObsTestUtils.DuplicateArray(new[] { 1f }, 4);
+            float[] expectedDefault = new float[] { 0f };
             GridObsTestUtils.AssertSubarraysAtIndex(output, subarrayIndicies, expectedSubarrays, expectedDefault);
         }
 
             Assert.AreEqual(10 * 10 * 1, output.Length);
 
             int[] subarrayIndicies = new int[] { 77, 78, 87, 88 };
-            float[][] expectedSubarrays = GridObsTestUtils.DuplicateArray(new float[] { .5f }, 4);
+            float[][] expectedSubarrays = GridObsTestUtils.DuplicateArray(new[] { .5f }, 4);
             float[] expectedDefault = new float[] { 0 };
             GridObsTestUtils.AssertSubarraysAtIndex(output, subarrayIndicies, expectedSubarrays, expectedDefault);
         }
             Assert.AreEqual(10 * 10 * 2, output.Length);
 
             int[] subarrayIndicies = new int[] { 77, 78, 87, 88 };
-            float[][] expectedSubarrays = GridObsTestUtils.DuplicateArray(new float[] { .5f, 1 }, 4);
+            float[][] expectedSubarrays = GridObsTestUtils.DuplicateArray(new[] { .5f, 1 }, 4);
             float[] expectedDefault = new float[] { 0, 0 };
             GridObsTestUtils.AssertSubarraysAtIndex(output, subarrayIndicies, expectedSubarrays, expectedDefault);
         }

com.unity.ml-agents.extensions/Tests/Editor/Sensors/CountingGridSensorShapeTests.cs

 using UnityEngine;
 using NUnit.Framework;
 using Unity.MLAgents.Extensions.Sensors;
-using NUnit.Framework.Internal;
 
 namespace Unity.MLAgents.Extensions.Tests.Sensors
 {

com.unity.ml-agents.extensions/Tests/Editor/Sensors/GridObservationPerceiveTests.cs

 using UnityEngine;
 using UnityEngine.TestTools;
 using Unity.MLAgents.Extensions.Sensors;
-using System.Linq;
 
 namespace Unity.MLAgents.Extensions.Tests.Sensors
 {
             Assert.AreEqual(10 * 10 * 1, output.Length);
 
             int[] subarrayIndicies = new int[] { 77, 78, 87, 88 };
-            float[][] expectedSubarrays = GridObsTestUtils.DuplicateArray(new float[] { 2f / 3f }, 4);
+            float[][] expectedSubarrays = GridObsTestUtils.DuplicateArray(new[] { 2f / 3f }, 4);
             float[] expectedDefault = new float[] { 0f };
             GridObsTestUtils.AssertSubarraysAtIndex(output, subarrayIndicies, expectedSubarrays, expectedDefault);
         }
             Assert.AreEqual(10 * 10 * 1, output.Length);
 
             int[] subarrayIndicies = new int[] { 77, 78, 87, 88 };
-            float[][] expectedSubarrays = GridObsTestUtils.DuplicateArray(new float[] { 1f / 3f }, 4);
+            float[][] expectedSubarrays = GridObsTestUtils.DuplicateArray(new[] { 1f / 3f }, 4);
             float[] expectedDefault = new float[] { 0f };
             GridObsTestUtils.AssertSubarraysAtIndex(output, subarrayIndicies, expectedSubarrays, expectedDefault);
         }

com.unity.ml-agents.extensions/Tests/Editor/Sensors/GridSensorTestUtils.cs

 using NUnit.Framework;
 using System;
-using System.Diagnostics;
 using System.Linq;
 
 namespace Unity.MLAgents.Extensions.Tests.Sensors
         /// </summary>
         /// <param name="arr">The array to convert to string</param>
         /// <param name="initialIndex">The initial index. Default 0</param>
-        /// <param name="numOfElements">The number of elements to print</param>
+        /// <param name="maxNumberOfElements">The number of elements to print</param>
         /// <returns>Human readable string</returns>
         public static string Array2Str<T>(T[] arr, int initialIndex = 0, int maxNumberOfElements = int.MaxValue)
         {
         /// If the total array is data from a 4x4x2 grid observation, total will be an array of size 32 and each sub array will have a size of 2.
         /// Let 3 cells at indicies (0, 1), (2, 2), and (3, 0) with values ([.1, .5]), ([.9, .7]), ([0, .2]), respectively.
         /// If the default values of cells are ([0, 0]) then the grid observation will be as follows:
-        /// [ [0,  0], [.1, .5], [ 0, 0 ], [0, 0], 
+        /// [ [0,  0], [.1, .5], [ 0, 0 ], [0, 0],
-        ///   
+        ///
-        /// 
+        ///
         /// The indicies of the activated cells in the flattened array will be 1, 10, and 12
         ///
         /// So to verify that the total array is as expected, AssertSubarraysAtIndex should be called as

com.unity.ml-agents.extensions/Tests/Editor/Sensors/PoseExtractorTests.cs

         {
             Assert.Throws<UnityAgentsException>(() =>
             {
-                var bad = new BadPoseExtractor();
+                var unused = new BadPoseExtractor();
             });
         }
 

com.unity.ml-agents.extensions/Tests/Editor/Sensors/RigidBodyPoseExtractorTests.cs

 using UnityEngine;
 using NUnit.Framework;
 using Unity.MLAgents.Extensions.Sensors;
-using UnityEditor;
 
 namespace Unity.MLAgents.Extensions.Tests.Sensors
 {
             Assert.AreEqual(0, poseExtractor.NumPoses);
 
             // Add an RB under the other GameObject. Constructor will find a rigid body, but not the root.
-            var otherRb = otherGameObj.AddComponent<Rigidbody>();
+            otherGameObj.AddComponent<Rigidbody>();
             poseExtractor = new RigidBodyPoseExtractor(rootRb, otherGameObj);
             Assert.AreEqual(0, poseExtractor.NumPoses);
         }
             var rb1 = rootObj.AddComponent<Rigidbody>();
 
             var go2 = new GameObject();
-            var rb2 = go2.AddComponent<Rigidbody>();
+            go2.AddComponent<Rigidbody>();
             go2.transform.SetParent(rootObj.transform);
 
             var joint = go2.AddComponent<ConfigurableJoint>();

com.unity.ml-agents.extensions/Tests/Editor/Sensors/RigidBodySensorTests.cs

         public static void CompareObservation(ISensor sensor, float[] expected)
         {
             string errorMessage;
-            bool isOK = SensorHelper.CompareObservation(sensor, expected, out errorMessage);
-            Assert.IsTrue(isOK, errorMessage);
+            bool isOk = SensorHelper.CompareObservation(sensor, expected, out errorMessage);
+            Assert.IsTrue(isOk, errorMessage);
+        }
+
+        public static void CompareObservation(ISensor sensor, float[,,] expected)
+        {
+            string errorMessage;
+            bool isOk = SensorHelper.CompareObservation(sensor, expected, out errorMessage);
+            Assert.IsTrue(isOk, errorMessage);
         }
     }
 

com.unity.ml-agents.extensions/Tests/Runtime/RuntimeExampleTest.cs

-using UnityEngine;
-using UnityEditor;
-using UnityEngine.TestTools;
-using System.Collections;
 
 namespace Unity.MLAgents.Extensions.Tests
 {

com.unity.ml-agents.extensions/package.json

   "unity": "2018.4",
   "description": "A source-only package for new features based on ML-Agents",
   "dependencies": {
-    "com.unity.ml-agents": "1.5.0-preview"
+    "com.unity.ml-agents": "1.8.0-preview"
   }
 }

com.unity.ml-agents/CHANGELOG.md

 and this project adheres to
 [Semantic Versioning](http://semver.org/spec/v2.0.0.html).
 
+## [Unreleased]
+### Major Changes
+#### com.unity.ml-agents (C#)
+#### ml-agents / ml-agents-envs / gym-unity (Python)
-## [Unreleased]
+### Minor Changes
+#### com.unity.ml-agents / com.unity.ml-agents.extensions (C#)
+#### ml-agents / ml-agents-envs / gym-unity (Python)
+- The `encoding_size` setting for RewardSignals has been deprecated. Please use `network_settings` instead. (#4982)
+### Bug Fixes
+#### com.unity.ml-agents (C#)
+#### ml-agents / ml-agents-envs / gym-unity (Python)
+- An issue that caused `GAIL` to fail for environments where agents can terminate episodes by self-sacrifice has been fixed. (#4971)
+
+## [1.8.0-preview] - 2021-02-17
+### Major Changes
+#### com.unity.ml-agents (C#)
+#### ml-agents / ml-agents-envs / gym-unity (Python)
+- TensorFlow trainers have been removed, please use the Torch trainers instead. (#4707)
+- A plugin system for `mlagents-learn` has been added. You can now define custom
+  `StatsWriter` implementations and register them to be called during training.
+  More types of plugins will be added in the future. (#4788)
+
+### Minor Changes
+#### com.unity.ml-agents / com.unity.ml-agents.extensions (C#)
+- The `ActionSpec` constructor is now public. Previously, it was not possible to create an
+  ActionSpec with both continuous and discrete actions from code. (#4896)
+- `StatAggregationMethod.Sum` can now be passed to `StatsRecorder.Add()`. This
+  will result in the values being summed (instead of averaged) when written to
+  TensorBoard. Thanks to @brccabral for the contribution! (#4816)
+- The upper limit for the time scale (by setting the `--time-scale` paramater in mlagents-learn) was
+  removed when training with a player. The Editor still requires it to be clamped to 100. (#4867)
+- Added the IHeuristicProvider interface to allow IActuators as well as Agent implement the Heuristic function to generate actions.
+  Updated the Basic example and the Match3 Example to use Actuators.
+  Changed the namespace and file names of classes in com.unity.ml-agents.extensions. (#4849)
+- Added `VectorSensor.AddObservation(IList<float>)`. `VectorSensor.AddObservation(IEnumerable<float>)`
+  is deprecated. The `IList` version is recommended, as it does not generate any
+  additional memory allocations. (#4887)
+- Added `ObservationWriter.AddList()` and deprecated `ObservationWriter.AddRange()`.
+  `AddList()` is recommended, as it does not generate any additional memory allocations. (#4887)
+- The Barracuda dependency was upgraded to 1.3.0. (#4898)
+- Added `ActuatorComponent.CreateActuators`, and deprecate `ActuatorComponent.CreateActuator`.  The
+  default implementation will wrap `ActuatorComponent.CreateActuator` in an array and return that. (#4899)
+- `InferenceDevice.Burst` was added, indicating that Agent's model will be run using Barracuda's Burst backend.
+  This is the default for new Agents, but existing ones that use `InferenceDevice.CPU` should update to
+  `InferenceDevice.Burst`. (#4925)
+- Add an InputActuatorComponent to allow the generation of Agent action spaces from an InputActionAsset.
+  Projects wanting to use this feature will need to add the
+  [Input System Package](https://docs.unity3d.com/Packages/com.unity.inputsystem@1.1/manual/index.html)
+  at version 1.1.0-preview.3 or later. (#4881)
+
+#### ml-agents / ml-agents-envs / gym-unity (Python)
+- Tensorboard now logs the Environment Reward as both a scalar and a histogram. (#4878)
+- Added a `--torch-device` commandline option to `mlagents-learn`, which sets the default
+  [`torch.device`](https://pytorch.org/docs/stable/tensor_attributes.html#torch.torch.device) used for training. (#4888)
+- The `--cpu` commandline option had no effect and was removed. Use `--torch-device=cpu` to force CPU training. (#4888)
+- The `mlagents_env` API has changed, `BehaviorSpec` now has a `observation_specs` property containing a list of `ObservationSpec`. For more information on `ObservationSpec` see [here](https://github.com/Unity-Technologies/ml-agents/blob/master/docs/Python-API.md#behaviorspec). (#4763, #4825)
+
+### Bug Fixes
+#### com.unity.ml-agents (C#)
+- Fix a compile warning about using an obsolete enum in `GrpcExtensions.cs`. (#4812)
+- CameraSensor now logs an error if the GraphicsDevice is null. (#4880)
+- Removed unnecessary memory allocations in `ActuatorManager.UpdateActionArray()` (#4877)
+- Removed unnecessary memory allocations in `SensorShapeValidator.ValidateSensors()` (#4879)
+- Removed unnecessary memory allocations in `SideChannelManager.GetSideChannelMessage()` (#4886)
+- Removed several memory allocations that happened during inference. On a test scene, this
+  reduced the amount of memory allocated by approximately 25%. (#4887)
+- Removed several memory allocations that happened during inference with discrete actions. (#4922)
+- Properly catch permission errors when writing timer files. (#4921)
+- Unexpected exceptions during training initialization and shutdown are now logged. If you see
+  "noisy" logs, please let us know! (#4930, #4935)
+
+#### ml-agents / ml-agents-envs / gym-unity (Python)
+- Fixed a bug that would cause an exception when `RunOptions` was deserialized via `pickle`. (#4842)
+- Fixed a bug that can cause a crash if a behavior can appear during training in multi-environment training. (#4872)
+- Fixed the computation of entropy for continuous actions. (#4869)
+- Fixed a bug that would cause `UnityEnvironment` to wait the full timeout
+  period and report a misleading error message if the executable crashed
+  without closing the connection. It now periodically checks the process status
+  while waiting for a connection, and raises a better error message if it crashes. (#4880)
+- Passing a `-logfile` option in the `--env-args` option to `mlagents-learn` is
+  no longer overwritten. (#4880)
+- The `load_weights` function was being called unnecessarily often in the Ghost Trainer leading to training slowdowns. (#4934)
+
+
+## [1.7.2-preview] - 2020-12-22
+### Bug Fixes
+#### com.unity.ml-agents (C#)
+- Add analytics package dependency to the package manifest. (#4794)
+#### ml-agents / ml-agents-envs / gym-unity (Python)
+- Fixed the docker build process. (#4791)
+
+## [1.7.0-preview] - 2020-12-21
+### Major Changes
+#### com.unity.ml-agents (C#)
+#### ml-agents / ml-agents-envs / gym-unity (Python)
+- PyTorch trainers now support training agents with both continuous and discrete action spaces. (#4702)
+The `.onnx` models generated by the trainers of this release are incompatible with versions of Barracuda before `1.2.1-preview`. If you upgrade the trainers, you must upgrade the version of the Barracuda package as well (which can be done by upgrading the `com.unity.ml-agents` package).
+### Minor Changes
+#### com.unity.ml-agents / com.unity.ml-agents.extensions (C#)
+- Agents with both continuous and discrete actions are now supported. You can specify
+both continuous and discrete action sizes in Behavior Parameters. (#4702, #4718)
+- In order to improve the developer experience for Unity ML-Agents Toolkit, we have added in-editor analytics.
+Please refer to "Information that is passively collected by Unity" in the
+[Unity Privacy Policy](https://unity3d.com/legal/privacy-policy). (#4677)
+- The FoodCollector example environment now uses continuous actions for moving and
+discrete actions for shooting. (#4746)
+#### ml-agents / ml-agents-envs / gym-unity (Python)
+- `ActionSpec.validate_action()` now enforces that `UnityEnvironment.set_action_for_agent()` receives a 1D `np.array`. (#4691)
+
+### Bug Fixes
+#### com.unity.ml-agents (C#)
+- Removed noisy warnings about API minor version mismatches in both the C# and python code. (#4688)
+#### ml-agents / ml-agents-envs / gym-unity (Python)
+
+
+## [1.6.0-preview] - 2020-11-18
 ### Major Changes
 #### com.unity.ml-agents (C#)
 #### ml-agents / ml-agents-envs / gym-unity (Python)
  adding `framework: tensorflow` in the configuration YAML. (#4517)
 
 ### Minor Changes
-#### com.unity.ml-agents (C#)
+#### com.unity.ml-agents / com.unity.ml-agents.extensions (C#)
+- Utilities were added to `com.unity.ml-agents.extensions` to make it easier to
+integrate with match-3 games. See the [readme](https://github.com/Unity-Technologies/ml-agents/blob/master/com.unity.ml-agents.extensions/Documentation~/Match3.md)
+for more details. (#4515)
 #### ml-agents / ml-agents-envs / gym-unity (Python)
 - The `action_probs` node is no longer listed as an output in TensorFlow models (#4613).
 
 Previously, this would result in an infinite loop and cause the editor to hang. (#4573)
 #### ml-agents / ml-agents-envs / gym-unity (Python)
 - Fixed an issue where runs could not be resumed when using TensorFlow and Ghost Training. (#4593)
+- Change the tensor type of step count from int32 to int64 to address the overflow issue when step
+goes larger than 2^31. Previous Tensorflow checkpoints will become incompatible and cannot be loaded. (#4607)
+- Remove extra period after "Training" in console log. (#4674)
 
 
 ## [1.5.0-preview] - 2020-10-14

com.unity.ml-agents/CONTRIBUTING.md

 
 ## Environments
 
-We are also actively open to adding community contributed environments as
-examples, as long as they are small, simple, demonstrate a unique feature of the
-platform, and provide a unique non-trivial challenge to modern machine learning
-algorithms. Feel free to submit these environments with a PR explaining the
-nature of the environment and task.
+We are currently not accepting environment contributions directly into ML-Agents.
+However, we believe community created enviornments have a lot of value to the
+community. If you have an interesting enviornment and are willing to share,
+feel free to showcase it and share any relevant files in the
+[ML-Agents forum](https://forum.unity.com/forums/ml-agents.453/).
 
 ## Continuous Integration (CI)
 

com.unity.ml-agents/Documentation~/com.unity.ml-agents.md

 # About ML-Agents package (`com.unity.ml-agents`)
 
-The Unity ML-Agents package contains the C# SDK for the [Unity ML-Agents
+The _ML-Agents_ package contains the primary C# SDK for the [Unity ML-Agents
 Toolkit].
 
 The package allows you to convert any Unity scene to into a learning environment
 instrumenting a Unity scene, setting it up for training, and then embedding the
 trained model back into your Unity scene. The machine learning algorithms that
 orchestrate training are part of the companion [Python package].
+
+Note that we also provide an _ML-Agents Extensions_ package
+(`com.unity.ml-agents.extensions`) that contains early/experimental features
+that you may find useful. This package is only available from the [ML-Agents
+GitHub repo].
 
 ## Package contents
 
 the documentation, you can checkout our [GitHub Repository], which also includes
 a number of ways to [connect with us] including our [ML-Agents Forum].
 
+In order to improve the developer experience for Unity ML-Agents Toolkit, we have added in-editor analytics.
+Please refer to "Information that is passively collected by Unity" in the
+[Unity Privacy Policy](https://unity3d.com/legal/privacy-policy).
+
-[installation instructions]: https://github.com/Unity-Technologies/ml-agents/blob/release_8_docs/docs/Installation.md
+[installation instructions]: https://github.com/Unity-Technologies/ml-agents/blob/release_12_docs/docs/Installation.md
+[ML-Agents GitHub repo]: https://github.com/Unity-Technologies/ml-agents/blob/release_12_docs/com.unity.ml-agents.extensions

com.unity.ml-agents/Editor/BehaviorParametersEditor.cs

 using Unity.MLAgents.Policies;
 using Unity.MLAgents.Sensors;
 using Unity.MLAgents.Sensors.Reflection;
-using UnityEngine;
 
 namespace Unity.MLAgents.Editor
 {
         float m_TimeSinceModelReload;
         // Whether or not the model needs to be reloaded
         bool m_RequireReload;
+        const string k_BehaviorName = "m_BehaviorName";
+        const string k_BrainParametersName = "m_BrainParameters";
+        const string k_ModelName = "m_Model";
+        const string k_InferenceDeviceName = "m_InferenceDevice";
+        const string k_BehaviorTypeName = "m_BehaviorType";
+        const string k_TeamIdName = "TeamId";
+        const string k_UseChildSensorsName = "m_UseChildSensors";
+        const string k_ObservableAttributeHandlingName = "m_ObservableAttributeHandling";
 
         public override void OnInspectorGUI()
         {
 
             EditorGUI.BeginChangeCheck();
             {
-                EditorGUILayout.PropertyField(so.FindProperty("m_BehaviorName"));
+                EditorGUILayout.PropertyField(so.FindProperty(k_BehaviorName));
+            EditorGUI.BeginChangeCheck();
-                EditorGUILayout.PropertyField(so.FindProperty("m_BrainParameters"), true);
+                EditorGUILayout.PropertyField(so.FindProperty(k_BrainParametersName), true);
-                EditorGUILayout.PropertyField(so.FindProperty("m_Model"), true);
+                EditorGUILayout.PropertyField(so.FindProperty(k_ModelName), true);
-                EditorGUILayout.PropertyField(so.FindProperty("m_InferenceDevice"), true);
+                EditorGUILayout.PropertyField(so.FindProperty(k_InferenceDeviceName), true);
                 EditorGUI.indentLevel--;
             }
             needPolicyUpdate = needPolicyUpdate || EditorGUI.EndChangeCheck();
-                EditorGUILayout.PropertyField(so.FindProperty("m_BehaviorType"));
+                EditorGUILayout.PropertyField(so.FindProperty(k_BehaviorTypeName));
-            EditorGUILayout.PropertyField(so.FindProperty("TeamId"));
+            EditorGUILayout.PropertyField(so.FindProperty(k_TeamIdName));
-                EditorGUILayout.PropertyField(so.FindProperty("m_UseChildSensors"), true);
-                EditorGUILayout.PropertyField(so.FindProperty("m_ObservableAttributeHandling"), true);
+                EditorGUILayout.PropertyField(so.FindProperty(k_UseChildSensorsName), true);
+                EditorGUILayout.PropertyField(so.FindProperty(k_ObservableAttributeHandlingName), true);
             }
             EditorGUI.EndDisabledGroup();
 
             // Display all failed checks
             D.logEnabled = false;
             Model barracudaModel = null;
-            var model = (NNModel)serializedObject.FindProperty("m_Model").objectReferenceValue;
+            var model = (NNModel)serializedObject.FindProperty(k_ModelName).objectReferenceValue;
-            SensorComponent[] sensorComponents;
-            if (behaviorParameters.UseChildSensors)
-            {
-                sensorComponents = behaviorParameters.GetComponentsInChildren<SensorComponent>();
-            }
-            else
-            {
-                sensorComponents = behaviorParameters.GetComponents<SensorComponent>();
-            }
+            var agent = behaviorParameters.gameObject.GetComponent<Agent>();
+            agent.sensors = new List<ISensor>();
+            agent.InitializeSensors();
+            var sensors = agent.sensors.ToArray();
 
             ActuatorComponent[] actuatorComponents;
             if (behaviorParameters.UseChildActuators)
             // Get the total size of the sensors generated by ObservableAttributes.
             // If there are any errors (e.g. unsupported type, write-only properties), display them too.
             int observableAttributeSensorTotalSize = 0;
-            var agent = behaviorParameters.GetComponent<Agent>();
             if (agent != null && behaviorParameters.ObservableAttributeHandling != ObservableAttributeOptions.Ignore)
             {
                 List<string> observableErrors = new List<string>();
             if (brainParameters != null)
             {
                 var failedChecks = Inference.BarracudaModelParamLoader.CheckModel(
-                    barracudaModel, brainParameters, sensorComponents, actuatorComponents,
+                    barracudaModel, brainParameters, sensors, actuatorComponents,
                     observableAttributeSensorTotalSize, behaviorParameters.BehaviorType
                 );
                 foreach (var check in failedChecks)

com.unity.ml-agents/Editor/BrainParametersDrawer.cs

         // The height of a line in the Unity Inspectors
         const float k_LineHeight = 17f;
         const int k_VecObsNumLine = 3;
-        const string k_ActionSizePropName = "VectorActionSize";
-        const string k_ActionTypePropName = "VectorActionSpaceType";
+        const string k_ActionSpecName = "m_ActionSpec";
+        const string k_ContinuousActionSizeName = "m_NumContinuousActions";
+        const string k_DiscreteBranchSizeName = "BranchSizes";
         const string k_ActionDescriptionPropName = "VectorActionDescriptions";
         const string k_VecObsPropName = "VectorObservationSize";
         const string k_NumVecObsPropName = "NumStackedVectorObservations";
         /// to make the custom GUI for.</param>
         static void DrawVectorAction(Rect position, SerializedProperty property)
         {
-            EditorGUI.LabelField(position, "Vector Action");
+            EditorGUI.LabelField(position, "Actions");
-            var bpVectorActionType = property.FindPropertyRelative(k_ActionTypePropName);
-            EditorGUI.PropertyField(
-                position,
-                bpVectorActionType,
-                new GUIContent("Space Type",
-                    "Corresponds to whether state vector contains  a single integer (Discrete) " +
-                    "or a series of real-valued floats (Continuous)."));
+            var actionSpecProperty = property.FindPropertyRelative(k_ActionSpecName);
+            DrawContinuousVectorAction(position, actionSpecProperty);
-            if (bpVectorActionType.enumValueIndex == 1)
-            {
-                DrawContinuousVectorAction(position, property);
-            }
-            else
-            {
-                DrawDiscreteVectorAction(position, property);
-            }
+            DrawDiscreteVectorAction(position, actionSpecProperty);
         }
 
         /// <summary>
         /// to make the custom GUI for.</param>
         static void DrawContinuousVectorAction(Rect position, SerializedProperty property)
         {
-            var vecActionSize = property.FindPropertyRelative(k_ActionSizePropName);
-
-            // This check is here due to:
-            // https://fogbugz.unity3d.com/f/cases/1246524/
-            // If this case has been resolved, please remove this if condition.
-            if (vecActionSize.arraySize != 1)
-            {
-                vecActionSize.arraySize = 1;
-            }
-            var continuousActionSize =
-                vecActionSize.GetArrayElementAtIndex(0);
+            var continuousActionSize = property.FindPropertyRelative(k_ContinuousActionSizeName);
-                new GUIContent("Space Size", "Length of continuous action vector."));
+                new GUIContent("Continuous Actions", "Number of continuous actions."));
         }
 
         /// <summary>
         /// to make the custom GUI for.</param>
         static void DrawDiscreteVectorAction(Rect position, SerializedProperty property)
         {
-            var vecActionSize = property.FindPropertyRelative(k_ActionSizePropName);
+            var branchSizes = property.FindPropertyRelative(k_DiscreteBranchSizeName);
-                position, "Branches Size", vecActionSize.arraySize);
+                position, "Discrete Branches", branchSizes.arraySize);
-            if (newSize != vecActionSize.arraySize)
+            if (newSize != branchSizes.arraySize)
-                vecActionSize.arraySize = newSize;
+                branchSizes.arraySize = newSize;
             }
 
             position.y += k_LineHeight;
-                 branchIndex < vecActionSize.arraySize;
+                 branchIndex < branchSizes.arraySize;
-                    vecActionSize.GetArrayElementAtIndex(branchIndex);
+                    branchSizes.GetArrayElementAtIndex(branchIndex);
 
                 EditorGUI.PropertyField(
                     position,
         /// <returns>The height of the drawer of the Vector Action.</returns>
         static float GetHeightDrawVectorAction(SerializedProperty property)
         {
-            var actionSize = 2 + property.FindPropertyRelative(k_ActionSizePropName).arraySize;
-            if (property.FindPropertyRelative(k_ActionTypePropName).enumValueIndex == 0)
-            {
-                actionSize += 1;
-            }
-            return actionSize * k_LineHeight;
+            var actionSpecProperty = property.FindPropertyRelative(k_ActionSpecName);
+            var numActionLines = 3 + actionSpecProperty.FindPropertyRelative(k_DiscreteBranchSizeName).arraySize;
+            return numActionLines * k_LineHeight;
         }
     }
 }

com.unity.ml-agents/Editor/CameraSensorComponentEditor.cs

-using UnityEngine;
 using UnityEditor;
 using Unity.MLAgents.Sensors;
 

com.unity.ml-agents/Editor/DemonstrationDrawer.cs

         SerializedProperty m_BrainParameters;
         SerializedProperty m_DemoMetaData;
         SerializedProperty m_ObservationShapes;
+        const string k_BrainParametersName = "brainParameters";
+        const string k_MetaDataName = "metaData";
+        const string k_ObservationSummariesName = "observationSummaries";
+        const string k_DemonstrationName = "demonstrationName";
+        const string k_NumberStepsName = "numberSteps";
+        const string k_NumberEpisodesName = "numberEpisodes";
+        const string k_MeanRewardName = "meanReward";
+        const string k_ActionSpecName = "m_ActionSpec";
+        const string k_NumContinuousActionsName = "m_NumContinuousActions";
+        const string k_NumDiscreteActionsName = "BranchSizes";
+        const string k_ShapeName = "shape";
+
-            m_BrainParameters = serializedObject.FindProperty("brainParameters");
-            m_DemoMetaData = serializedObject.FindProperty("metaData");
-            m_ObservationShapes = serializedObject.FindProperty("observationSummaries");
+            m_BrainParameters = serializedObject.FindProperty(k_BrainParametersName);
+            m_DemoMetaData = serializedObject.FindProperty(k_MetaDataName);
+            m_ObservationShapes = serializedObject.FindProperty(k_ObservationSummariesName);
         }
 
         /// <summary>
         {
-            var nameProp = property.FindPropertyRelative("demonstrationName");
-            var experiencesProp = property.FindPropertyRelative("numberSteps");
-            var episodesProp = property.FindPropertyRelative("numberEpisodes");
-            var rewardsProp = property.FindPropertyRelative("meanReward");
+            var nameProp = property.FindPropertyRelative(k_DemonstrationName);
+            var experiencesProp = property.FindPropertyRelative(k_NumberStepsName);
+            var episodesProp = property.FindPropertyRelative(k_NumberEpisodesName);
+            var rewardsProp = property.FindPropertyRelative(k_MeanRewardName);
 
             var nameLabel = nameProp.displayName + ": " + nameProp.stringValue;
             var experiencesLabel = experiencesProp.displayName + ": " + experiencesProp.intValue;
         /// </summary>
         void MakeActionsProperty(SerializedProperty property)
         {
-            var actSizeProperty = property.FindPropertyRelative("VectorActionSize");
-            var actSpaceTypeProp = property.FindPropertyRelative("VectorActionSpaceType");
-
-            var vecActSizeLabel =
-                actSizeProperty.displayName + ": " + BuildIntArrayLabel(actSizeProperty);
-            var actSpaceTypeLabel = actSpaceTypeProp.displayName + ": " +
-                (SpaceType)actSpaceTypeProp.enumValueIndex;
-
-            EditorGUILayout.LabelField(vecActSizeLabel);
-            EditorGUILayout.LabelField(actSpaceTypeLabel);
+            var actSpecProperty = property.FindPropertyRelative(k_ActionSpecName);
+            var continuousSizeProperty = actSpecProperty.FindPropertyRelative(k_NumContinuousActionsName);
+            var discreteSizeProperty = actSpecProperty.FindPropertyRelative(k_NumDiscreteActionsName);
+            var continuousSizeLabel = "Continuous Actions: " + continuousSizeProperty.intValue;
+            var discreteSizeLabel = "Discrete Action Branches: ";
+            discreteSizeLabel += discreteSizeProperty == null ? "[]" : BuildIntArrayLabel(discreteSizeProperty);
+            EditorGUILayout.LabelField(continuousSizeLabel);
+            EditorGUILayout.LabelField(discreteSizeLabel);
         }
 
         /// <summary>
             for (var i = 0; i < numObservations; i++)
             {
                 var summary = obsSummariesProperty.GetArrayElementAtIndex(i);
-                var shapeProperty = summary.FindPropertyRelative("shape");
+                var shapeProperty = summary.FindPropertyRelative(k_ShapeName);
                 shapesLabels.Add(BuildIntArrayLabel(shapeProperty));
             }
 

com.unity.ml-agents/Editor/RenderTextureSensorComponentEditor.cs

-using UnityEngine;
 using UnityEditor;
 using Unity.MLAgents.Sensors;
 namespace Unity.MLAgents.Editor

com.unity.ml-agents/Runtime/Academy.cs

  * API. For more information on each of these entities, in addition to how to
  * set-up a learning environment and train the behavior of characters in a
  * Unity scene, please browse our documentation pages on GitHub:
- * https://github.com/Unity-Technologies/ml-agents/tree/release_8_docs/docs/
+ * https://github.com/Unity-Technologies/ml-agents/tree/release_12_docs/docs/
  */
 
 namespace Unity.MLAgents
     /// fall back to inference or heuristic decisions. (You can also set agents to always use
     /// inference or heuristics.)
     /// </remarks>
-    [HelpURL("https://github.com/Unity-Technologies/ml-agents/tree/release_8_docs/" +
+    [HelpURL("https://github.com/Unity-Technologies/ml-agents/tree/release_12_docs/" +
         "docs/Learning-Environment-Design.md")]
     public class Academy : IDisposable
     {
         ///         <term>1.2.0</term>
         ///         <description>Support compression mapping for stacked compressed observations.</description>
         ///     </item>
+        ///     <item>
+        ///         <term>1.3.0</term>
+        ///         <description>Support both continuous and discrete actions.</description>
+        ///     </item>
+        ///     <item>
+        ///         <term>1.4.0</term>
+        ///         <description>Support training analytics sent from python trainer to the editor.</description>
+        ///     </item>
+        ///     <item>
+        ///         <term>1.5.0</term>
+        ///         <description>Support variable length observation training.</description>
+        ///     </item>
-        const string k_ApiVersion = "1.2.0";
+        const string k_ApiVersion = "1.5.0";
-        internal const string k_PackageVersion = "1.5.0-preview";
+        internal const string k_PackageVersion = "1.8.0-preview";
 
         const int k_EditorTrainingPort = 5004;
 
                 Dispose();
             }
         }
+
 #endif
 
         /// <summary>
             EnableAutomaticStepping();
 
             SideChannelManager.RegisterSideChannel(new EngineConfigurationChannel());
+            SideChannelManager.RegisterSideChannel(new TrainingAnalyticsSideChannel());
             m_EnvironmentParameters = new EnvironmentParameters();
             m_StatsRecorder = new StatsRecorder();
 
             {
-                Communicator = new RpcCommunicator(
-                    new CommunicatorInitParameters
-                    {
-                        port = port
-                    }
-                );
+                Communicator = CommunicatorFactory.Create();
             }
 
             if (Communicator != null)
-                //environment must use Inference.
+                // environment must use Inference.
+                bool initSuccessful = false;
+                var communicatorInitParams = new CommunicatorInitParameters
+                {
+                    port = port,
+                    unityCommunicationVersion = k_ApiVersion,
+                    unityPackageVersion = k_PackageVersion,
+                    name = "AcademySingleton",
+                    CSharpCapabilities = new UnityRLCapabilities()
+                };
+
-                    var unityRlInitParameters = Communicator.Initialize(
-                        new CommunicatorInitParameters
-                        {
-                            unityCommunicationVersion = k_ApiVersion,
-                            unityPackageVersion = k_PackageVersion,
-                            name = "AcademySingleton",
-                            CSharpCapabilities = new UnityRLCapabilities()
-                        });
-                    UnityEngine.Random.InitState(unityRlInitParameters.seed);
-                    // We might have inference-only Agents, so set the seed for them too.
-                    m_InferenceSeed = unityRlInitParameters.seed;
-                    TrainerCapabilities = unityRlInitParameters.TrainerCapabilities;
-                    TrainerCapabilities.WarnOnPythonMissingBaseRLCapabilities();
+                    initSuccessful = Communicator.Initialize(
+                        communicatorInitParams,
+                        out var unityRlInitParameters
+                    );
+                    if (initSuccessful)
+                    {
+                        UnityEngine.Random.InitState(unityRlInitParameters.seed);
+                        // We might have inference-only Agents, so set the seed for them too.
+                        m_InferenceSeed = unityRlInitParameters.seed;
+                        TrainerCapabilities = unityRlInitParameters.TrainerCapabilities;
+                        TrainerCapabilities.WarnOnPythonMissingBaseRLCapabilities();
+                    }
+                    else
+                    {
+                        Debug.Log($"Couldn't connect to trainer on port {port} using API version {k_ApiVersion}. Will perform inference instead.");
+                        Communicator = null;
+                    }
-                catch
+                catch (Exception ex)
-                    Debug.Log($"" +
-                        $"Couldn't connect to trainer on port {port} using API version {k_ApiVersion}. " +
-                        "Will perform inference instead."
-                    );
+                    Debug.Log($"Unexpected exception when trying to initialize communication: {ex}\nWill perform inference instead.");
+
             if (Communicator != null)
             {
                 Communicator.QuitCommandReceived += OnQuitCommandReceived;
         /// NNModel and the InferenceDevice as provided.
         /// </summary>
         /// <param name="model">The NNModel the ModelRunner must use.</param>
-        /// <param name="actionSpec"> Description of the action spaces for the Agent.</param>
+        /// <param name="actionSpec"> Description of the actions for the Agent.</param>
         /// <param name="inferenceDevice">
         /// The inference device (CPU or GPU) the ModelRunner will use.
         /// </param>

com.unity.ml-agents/Runtime/Actuators/ActionSegment.cs

             System.Array.Clear(Array, Offset, Length);
         }
 
+        /// <summary>
+        /// Check if the segment is empty.
+        /// </summary>
+        /// <returns>Whether or not the segment is empty.</returns>
+        public bool IsEmpty()
+        {
+            return Array == null || Array.Length == 0;
+        }
+
         /// <inheritdoc/>
         IEnumerator<T> IEnumerable<T>.GetEnumerator()
         {

com.unity.ml-agents/Runtime/Actuators/ActionSpec.cs

 using System;
 using System.Collections.Generic;
 using System.Linq;
-using Unity.MLAgents.Policies;
+using UnityEngine;
-    /// Defines the structure of an Action Space to be used by the Actuator system.
+    /// Defines the structure of the actions to be used by the Actuator system.
-    public readonly struct ActionSpec
+    [Serializable]
+    public struct ActionSpec
+        [SerializeField]
+        int m_NumContinuousActions;
-        /// An array of branch sizes for our action space.
+        /// An array of branch sizes for discrete actions.
-        /// For an IActuator that uses a Discrete <see cref="SpaceType"/>, the number of
+        /// For an IActuator that uses discrete actions, the number of
         /// branches is the Length of the Array and each index contains the branch size.
         /// The cumulative sum of the total number of discrete actions can be retrieved
         /// by the <see cref="SumOfDiscreteBranchSizes"/> property.
-        public readonly int[] BranchSizes;
+        public int[] BranchSizes;
-        /// The number of actions for a Continuous <see cref="SpaceType"/>.
+        /// The number of continuous actions that an Agent can take.
-        public int NumContinuousActions { get; }
+        public int NumContinuousActions { get { return m_NumContinuousActions; } set { m_NumContinuousActions = value; } }
-        /// The number of branches for a Discrete <see cref="SpaceType"/>.
+        /// The number of branches for discrete actions that an Agent can take.
-        public int NumDiscreteActions { get; }
+        public int NumDiscreteActions { get { return BranchSizes == null ? 0 : BranchSizes.Length; } }
-        public int SumOfDiscreteBranchSizes { get; }
+        public int SumOfDiscreteBranchSizes { get { return BranchSizes == null ? 0 : BranchSizes.Sum(); } }
-        /// <param name="numActions">The number of actions available.</param>
+        /// <param name="numActions">The number of continuous actions available.</param>
-            var actuatorSpace = new ActionSpec(numActions, 0);
+            var actuatorSpace = new ActionSpec(numActions, null);
             return actuatorSpace;
         }
 
         /// </summary>
-        /// <param name="branchSizes">The array of branch sizes for the discrete action space.  Each index
+        /// <param name="branchSizes">The array of branch sizes for the discrete actions.  Each index
-            var numActions = branchSizes.Length;
-            var actuatorSpace = new ActionSpec(0, numActions, branchSizes);
+            var actuatorSpace = new ActionSpec(0, branchSizes);
-        internal ActionSpec(int numContinuousActions, int numDiscreteActions, int[] branchSizes = null)
+        /// <summary>
+        /// Create an ActionSpec initialized with the specified action sizes.
+        /// </summary>
+        /// <param name="numContinuousActions">The number of continuous actions available.</param>
+        /// <param name="discreteBranchSizes">The array of branch sizes for the discrete actions.  Each index
+        /// contains the number of actions available for that branch.</param>
+        /// <returns>An ActionSpec initialized with the specified action sizes.</returns>
+        public ActionSpec(int numContinuousActions = 0, int[] discreteBranchSizes = null)
-            NumContinuousActions = numContinuousActions;
-            NumDiscreteActions = numDiscreteActions;
-            BranchSizes = branchSizes;
-            SumOfDiscreteBranchSizes = branchSizes?.Sum() ?? 0;
+            m_NumContinuousActions = numContinuousActions;
+            BranchSizes = discreteBranchSizes ?? Array.Empty<int>();
-        /// Temporary check that the ActionSpec uses either all continuous or all discrete actions.
-        /// This should be removed once the trainer supports them.
+        /// Check that the ActionSpec uses either all continuous or all discrete actions.
+        /// This is only used when connecting to old versions of the trainer that don't support this.
-        internal void CheckNotHybrid()
+        internal void CheckAllContinuousOrDiscrete()
-                throw new UnityAgentsException("ActionSpecs must be all continuous or all discrete.");
+                throw new UnityAgentsException(
+                    "Action spaces with both continuous and discrete actions are not supported by the trainer. " +
+                    "ActionSpecs must be all continuous or all discrete."
+                );
+        }
+
+        /// <summary>
+        /// Combines a list of actions specs and allocates a new array of branch sizes if needed.
+        /// </summary>
+        /// <param name="specs">The list of action specs to combine.</param>
+        /// <returns>An ActionSpec which represents the aggregate of the ActionSpecs passed in.</returns>
+        public static ActionSpec Combine(params ActionSpec[] specs)
+        {
+            var numContinuous = 0;
+            var numDiscrete = 0;
+            for (var i = 0; i < specs.Length; i++)
+            {
+                var spec = specs[i];
+                numContinuous += spec.NumContinuousActions;
+                numDiscrete += spec.NumDiscreteActions;
+            }
+
+            if (numDiscrete <= 0)
+            {
+                return MakeContinuous(numContinuous);
+            }
+
+            var branchSizes = new int[numDiscrete];
+            var offset = 0;
+            for (var i = 0; i < specs.Length; i++)
+            {
+                var spec = specs[i];
+                if (spec.BranchSizes.Length == 0)
+                {
+                    continue;
+                }
+                var branchSizesLength = spec.BranchSizes.Length;
+                Array.Copy(spec.BranchSizes,
+                    0,
+                    branchSizes,
+                    offset,
+                    branchSizesLength);
+                offset += branchSizesLength;
+            }
+            return new ActionSpec(numContinuous, branchSizes);
         }
     }
 }

com.unity.ml-agents/Runtime/Actuators/ActuatorComponent.cs

+using System;
 using UnityEngine;
 
 namespace Unity.MLAgents.Actuators
         /// Create the IActuator.  This is called by the Agent when it is initialized.
         /// </summary>
         /// <returns>Created IActuator object.</returns>
+        [Obsolete("Use CreateActuators instead.")]
-        /// The specification of the Action space for this ActuatorComponent.
+        /// Create a collection of <see cref="IActuator"/>s.  This is called by the <see cref="Agent"/> during
+        /// initialization.
+        /// </summary>
+        /// <returns>A collection of <see cref="IActuator"/>s</returns>
+        public virtual IActuator[] CreateActuators()
+        {
+#pragma warning disable 618
+            return new[] { CreateActuator() };
+#pragma warning restore 618
+        }
+
+        /// <summary>
+        /// The specification of the possible actions for this ActuatorComponent.
         /// This must produce the same results as the corresponding IActuator's ActionSpec.
         /// </summary>
         /// <seealso cref="ActionSpec"/>