Skip to content

Add use curiosity in Hallway example environment #1952

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Merged
merged 2 commits into from
Apr 16, 2019
Merged

Add use curiosity in Hallway example environment #1952

merged 2 commits into from
Apr 16, 2019

Conversation

sdiao
Copy link
Contributor

@sdiao sdiao commented Apr 16, 2019

Hallway example takes a long time to train. To speed up training, we can enable curiosity. This change is to add a suggestion in the doc in case our customers want to speed up the Hallway example training.

@sdiao sdiao requested a review from harperj April 16, 2019 21:01
@sdiao sdiao changed the base branch from master to develop April 16, 2019 21:02
mmattar
mmattar reviewed Apr 16, 2019
View reviewed changes
docs/Learning-Environment-Examples.md
@@ -246,6 +246,7 @@ If you would like to contribute environments, please see our
`VisualHallway` scene.
* Reset Parameters: None.
* Benchmark Mean Reward: 0.7
* To speed up training, you can enable curiosity by adding `use_curiosity: true` in `config/trainer_config.yaml`
Copy link

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Is the intention to have space prior to the asterisk?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

yes. It looks all examples have the same list of bullet points (Set-up, Goal,..., Benchmark Mean reward). I want to keep the overall structure, and only add enabling curiosity to achieve the benchmark mean reward faster. So I add this optimization under the benchmark mean reward. Let me know if there is a better place to add a note of enabling curiosity.

@mmattar mmattar merged commit 45c535b into develop Apr 16, 2019
mantasp added a commit that referenced this pull request Jun 14, 2019
@mantasp
Merge branch 'develop-barracuda-0.2.0' into develop-barracuda-test
c9551d8 
* develop-barracuda-0.2.0: (80 commits)
  Restored compatibility fixes for some Androids
  Upgraded to Barracuda 0.2.1, fixes issues with discrete action models
  First stage of ML Agents update to Barracuda 0.2.x
  Update Learning-Environment-Create-New.md (#1993)
  Format gym_unity using black
  [Documentation] Added information for the environments the trainer cannot train with the default configurations (#1995)
  [Documentation] SetReward method (#1996)
  [Gym] Added no_graphics argument (#1997)
  Develop black format fix (#1998)
  Add exception for external brains and array-ify (#1971)
  Enable the switching of scene to control mode, load the corresponding scene using environment variable (#1956)
  Add use curiosity in Hallway example environment (#1952)
  Python code reformat  via [`black`](https://github.com/ambv/black). Features: - Reformat code via black. - Adding circleci configurations. - Add contribution guidelines.
  Fix custom protobuf link in README
  Bumping gym_unity version
  Release 0.8.1 to fix pypi issues
  Explicitly adding all packages as imports
  Migration doc fixes
  Basic Retrain (#1929)
  Updated the 3dballhard model
  ...

# Conflicts:
#	UnitySDK/Assets/ML-Agents/Examples/3DBall/TFModels/3DBallHardLearning.nn
#	UnitySDK/Assets/ML-Agents/Examples/3DBall/TFModels/3DBallHardLearning.nn.meta
#	UnitySDK/Assets/ML-Agents/Examples/3DBall/TFModels/3DBallLearning.bytes.meta
#	UnitySDK/Assets/ML-Agents/Examples/3DBall/TFModels/3DBallLearning.nn
#	UnitySDK/Assets/ML-Agents/Examples/3DBall/TFModels/3DBallLearning.nn.meta
#	UnitySDK/Assets/ML-Agents/Examples/BananaCollectors/TFModels/BananaLearning.nn
#	UnitySDK/Assets/ML-Agents/Examples/BananaCollectors/TFModels/BananaLearning.nn.meta
#	UnitySDK/Assets/ML-Agents/Examples/Basic/Brains/BasicLearning.asset
#	UnitySDK/Assets/ML-Agents/Examples/Basic/TFModels/BasicLearning.bytes.meta
#	UnitySDK/Assets/ML-Agents/Examples/Basic/TFModels/BasicLearning.nn
#	UnitySDK/Assets/ML-Agents/Examples/Basic/TFModels/BasicLearning.nn.meta
#	UnitySDK/Assets/ML-Agents/Examples/Bouncer/TFModels/BouncerLearning.nn
#	UnitySDK/Assets/ML-Agents/Examples/Bouncer/TFModels/BouncerLearning.nn.meta
#	UnitySDK/Assets/ML-Agents/Examples/Crawler/TFModels/CrawlerDynamicLearning.nn
#	UnitySDK/Assets/ML-Agents/Examples/Crawler/TFModels/CrawlerDynamicLearning.nn.meta
#	UnitySDK/Assets/ML-Agents/Examples/Crawler/TFModels/CrawlerStaticLearning.bytes.meta
#	UnitySDK/Assets/ML-Agents/Examples/Crawler/TFModels/CrawlerStaticLearning.nn
#	UnitySDK/Assets/ML-Agents/Examples/Crawler/TFModels/CrawlerStaticLearning.nn.meta
#	UnitySDK/Assets/ML-Agents/Examples/GridWorld/Scenes/GridWorld.unity
#	UnitySDK/Assets/ML-Agents/Examples/GridWorld/TFModels/GridWorldLearning.nn
#	UnitySDK/Assets/ML-Agents/Examples/GridWorld/TFModels/GridWorldLearning.nn.meta
#	UnitySDK/Assets/ML-Agents/Examples/Hallway/TFModels/HallwayLearning.nn
#	UnitySDK/Assets/ML-Agents/Examples/Hallway/TFModels/HallwayLearning.nn.meta
#	UnitySDK/Assets/ML-Agents/Examples/PushBlock/TFModels/PushBlockLearning.nn
#	UnitySDK/Assets/ML-Agents/Examples/PushBlock/TFModels/PushBlockLearning.nn.meta
#	UnitySDK/Assets/ML-Agents/Examples/Pyramids/TFModels/PyramidsLearning.nn
#	UnitySDK/Assets/ML-Agents/Examples/Pyramids/TFModels/PyramidsLearning.nn.meta
#	UnitySDK/Assets/ML-Agents/Examples/Reacher/TFModels/ReacherLearning.nn
#	UnitySDK/Assets/ML-Agents/Examples/Reacher/TFModels/ReacherLearning.nn.meta
#	UnitySDK/Assets/ML-Agents/Examples/Soccer/TFModels/GoalieLearning.nn
#	UnitySDK/Assets/ML-Agents/Examples/Soccer/TFModels/GoalieLearning.nn.meta
#	UnitySDK/Assets/ML-Agents/Examples/Soccer/TFModels/StrikerLearning.nn
#	UnitySDK/Assets/ML-Agents/Examples/Soccer/TFModels/StrikerLearning.nn.meta
#	UnitySDK/Assets/ML-Agents/Examples/Tennis/Brains/TennisLearning.asset
#	UnitySDK/Assets/ML-Agents/Examples/Tennis/TFModels/TennisLearning.nn
#	UnitySDK/Assets/ML-Agents/Examples/Tennis/TFModels/TennisLearning.nn.meta
#	UnitySDK/Assets/ML-Agents/Examples/Walker/TFModels/WalkerLearning.nn
#	UnitySDK/Assets/ML-Agents/Examples/Walker/TFModels/WalkerLearning.nn.meta
#	UnitySDK/Assets/ML-Agents/Examples/WallJump/TFModels/BigWallJumpLearning.nn
#	UnitySDK/Assets/ML-Agents/Examples/WallJump/TFModels/BigWallJumpLearning.nn.meta
#	UnitySDK/Assets/ML-Agents/Examples/WallJump/TFModels/SmallWallJumpLearning.nn
#	UnitySDK/Assets/ML-Agents/Examples/WallJump/TFModels/SmallWallJumpLearning.nn.meta
#	UnitySDK/Assets/ML-Agents/Plugins/Barracuda.Core/Barracuda.md
#	UnitySDK/Assets/ML-Agents/Plugins/Barracuda.Core/Barracuda/Barracuda.dll
#	UnitySDK/Assets/ML-Agents/Plugins/Barracuda.Core/Barracuda/Resources/Activation.compute
#	UnitySDK/Assets/ML-Agents/Plugins/Barracuda.Core/Barracuda/Resources/BarracudaReferenceImpl.compute
#	UnitySDK/Assets/ML-Agents/Plugins/Barracuda.Core/Barracuda/Resources/Broadcast.compute
#	UnitySDK/Assets/ML-Agents/Plugins/Barracuda.Core/Barracuda/Resources/Conv.compute
#	UnitySDK/Assets/ML-Agents/Plugins/Barracuda.Core/Barracuda/Resources/ConvOld.compute
#	UnitySDK/Assets/ML-Agents/Plugins/Barracuda.Core/Barracuda/Resources/Dense.compute
#	UnitySDK/Assets/ML-Agents/Plugins/Barracuda.Core/Barracuda/Resources/DenseFP16.compute
#	UnitySDK/Assets/ML-Agents/Plugins/Barracuda.Core/Barracuda/Resources/Experimental.compute
#	UnitySDK/Assets/ML-Agents/Plugins/Barracuda.Core/Barracuda/Resources/FastNV.compute
#	UnitySDK/Assets/ML-Agents/Plugins/Barracuda.Core/Barracuda/Resources/Generic.compute
#	UnitySDK/Assets/ML-Agents/Plugins/Barracuda.Core/Barracuda/Resources/Random.cginc
#	UnitySDK/Assets/ML-Agents/Plugins/Barracuda.Core/Barracuda/Resources/Tensor.cginc
#	UnitySDK/Assets/ML-Agents/Plugins/Barracuda.Core/Barracuda/Resources/TexConv.compute
#	UnitySDK/Assets/ML-Agents/Plugins/Barracuda.Core/LICENSE.md.meta
#	UnitySDK/Assets/ML-Agents/Plugins/Barracuda.Core/ReleaseNotes.md
#	UnitySDK/Assets/ML-Agents/Plugins/Barracuda.Core/ReleaseNotes.md.meta
#	UnitySDK/Assets/ML-Agents/Plugins/Barracuda.Core/package.json
#	UnitySDK/Assets/ML-Agents/Resources/NNModelIcon.png
#	UnitySDK/Assets/ML-Agents/Resources/NNModelIcon.png.meta
#	UnitySDK/Assets/ML-Agents/Scripts/Academy.cs
#	UnitySDK/Assets/ML-Agents/Scripts/InferenceBrain/BarracudaModelParamLoader.cs
#	UnitySDK/Assets/ML-Agents/Scripts/InferenceBrain/GeneratorImpl.cs
#	UnitySDK/Assets/ML-Agents/Scripts/LearningBrain.cs
#	UnitySDK/ProjectSettings/ProjectSettings.asset
#	docs/Basic-Guide.md
#	docs/Unity-Inference-Engine.md
#	docs/Using-TensorFlow-Sharp-in-Unity.md
#	ml-agents/mlagents/trainers/barracuda.py
#	ml-agents/mlagents/trainers/policy.py
#	ml-agents/mlagents/trainers/tensorflow_to_barracuda.py
@awjuliani awjuliani deleted the develop-sdiao-update-example-env branch July 23, 2019 20:18
LeSphax pushed a commit to LeSphax/ml-agents-1 that referenced this pull request May 3, 2020
@sdiao
Add use curiosity in Hallway example environment (Unity-Technologies#…
771bf34 
…1952)
@github-actions github-actions bot locked as resolved and limited conversation to collaborators May 18, 2021
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Reviewers

@mmattar mmattar mmattar left review comments

@harperj harperj Awaiting requested review from harperj

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
3 participants
@sdiao @mmattar @eshvk