ml-agents

作者	SHA1	备注	提交日期
Arthur Juliani	982fab41	Initial commit	7 年前
vincentpierre	cde3c8f7	formating and added documentation	7 年前
Alexander Scheurer	3152c971	--keep-checkpoints=<n> option for ppo.py	7 年前
Arthur Juliani	71591043	PPO additions and warnings * Add linear decay to learning rate for PPO * Add warning/exception for unsupported brain configurations w/ PPO	7 年前
GitHub	aee5d336	Fix discrete state (#33 ) * made BrainParameters a class to set default values Modified the error message if the state is discrete * Add discrete state support to PPO and provide discrete state example environment * Add flexibility to continuous control as well * Finish PPO flexible model generation implementation * Fix formatting * Support color observations * Add best practices document * bug fix for non square observations * Update Readme.md * Remove scipy dependency * Add installation doc	7 年前
Arthur Juliani	cd3bfb87	Added worker-id flag and pass through to enviroment in order to more easily manage multiple running simulations. (#40 )	7 年前
Arthur Juliani	b6ce30bf	Add curriculum support to PPO	7 年前
Arthur Juliani	e6696ed3	Don't print	7 年前
Arthur Juliani	4a11c005	Add curriculum code to notebook and simplify	7 年前
Arthur Juliani	06d9bbec	Log lesson in TensorBoard	7 年前
vincentpierre	2b8353b2	porting the changes on ppo.py and removing AgentMonitor.cs	7 年前
vincentpierre	d71ee998	changes on the ppo.py	7 年前
vincentpierre	6e950cd3	Can now switch inference configuration on/off in the editor. Reintroduced the broadcast feature for the non-External brains. Introduced the API number to check the compatibility between Unity and Python.	7 年前
Arthur Juliani	d1b81a32	Add push curriculum	7 年前
Arthur Juliani	51f23cd2	0.2 Update * added broadcast to the player and heuristic brain. Allows the python API to record actions taken along with the states and rewards * removed the broadcast checkbox Added a Handshake method for the communicator The academy will try to handshake regardless of the brains present Player and Heuristic brains will send their information through the communicator but will not receive commands * bug fix : The environment only requests actions from external brains when unique * added warning in case no brins are set to external * fix on the instanciation of coreBrains, fix on the conversion of actions to arrays in the BrainInfo received from step * default discrete action is now 0 bug fix for discrete broadcast action (the action size should be one in Agents.cs) modified Tennis so that the default action is no action modified the TemplateDecsion.cs to ensure non null values are sent from Decide() and MakeMemory() * minor fixes * need to convert the s...	7 年前
Arthur Juliani	b56259f6	Fix cumulative reward (Unity) and Nan reward (python) bugs	7 年前
Arthur Juliani	216888ee	Fixed to give lesson index parameter when start up (#179 ) * fixed to give lesson parameter when start up * applied to PPO.ipynb and modified ppo.py a bit	7 年前
vincentpierre	ebaf5268	ignoring the Packages folder that is created in unity-environment for Unity version 2017.3 Print a message if someone tries to lauch ppo with the load flag but an invalid run-path	7 年前
Arthur Juliani	75ea16ff	Add comments and alphabetize flags	7 年前
Arthur Juliani	fc1b8a1b	Fix academy reset out of order	7 年前
Arthur Juliani	ce2ce437	Added growth parameter to stop failing with allocation under windows for #277 (#278 )	7 年前
vincentpierre	d01bd6c2	the bytes file will besaved under the name of the environment, not its path	7 年前
Arthur Juliani	54652c69	dev-logParam (#135 ) * added the method write text to trainer so it is easy to write log the hyperparameters as a dictionary. Note: needs tensorflow version r1.2 or above * added message if impossible to write text summary in Tensorboard	7 年前
Arthur Juliani	94c20ef0	Curriculum documentation and improved Area code	7 年前

24 次代码提交 (43ac4148-babe-4d2b-954a-8c3c1cc609ce)