Search Unity

Search Results

Search Again
Page 1 of 4
  1. awjuliani
    Post

    Issues training with raytracing

    Hello. Unfortunately I don't have a device which supports ray-tracing handy to test this. Just to confirm, you are able to use the ML-Agents...
    Post by: awjuliani, Mar 16, 2021 in forum: ML-Agents
  2. awjuliani
    Post

    Runs don't stop at max_steps in inference mode?

    Unfortunately we currently do not support ending training in inference mode using the max_steps. Your idea of setting the learning rate to zero...
    Post by: awjuliani, Mar 16, 2021 in forum: ML-Agents
  3. awjuliani
    Post

    Question keep_checkpoints

    The keep_checkpoint has no impact on training performance. It is used to determine how many old models are saves during the training process. See...
    Post by: awjuliani, Mar 16, 2021 in forum: ML-Agents
  4. awjuliani
    Post

    Resolved Train Nao robot to walk

    Can you share more about what you modified from Walker to implement your agent? It could be possible that the body of your robot does not...
    Post by: awjuliani, Mar 15, 2021 in forum: ML-Agents
  5. awjuliani
    Post

    Why is the STD of mean reward so high?

    Hello. Yes, a high standard deviation corresponds to the agent having a variety of different final rewards in the training episodes. For tasks...
    Post by: awjuliani, Mar 15, 2021 in forum: ML-Agents
  6. awjuliani
    Post

    How to set decision-period from the python side when training with env as executable?

    It looks like the DecisionPeriod is indeed a public field on the requester, so you should be able to directly modify it:...
    Post by: awjuliani, Mar 12, 2021 in forum: ML-Agents
  7. awjuliani
    Post

    Smart Karts Training - mlagents_envs.exception.UnityTimeOutException: The Unity environment took too

    Hi. As per the message, it may be the case that the example you are running is not compatible with later versions of ML-Agents. I would recommend...
    Post by: awjuliani, Mar 12, 2021 in forum: ML-Agents
  8. awjuliani
    Post

    How to set decision-period from the python side when training with env as executable?

    You should be able to store the received float information in a public variable in your SideChannel class, and then query it by your agent, if the...
    Post by: awjuliani, Mar 11, 2021 in forum: ML-Agents
  9. awjuliani
    Post

    How to set decision-period from the python side when training with env as executable?

    Hello, Since agents can call RequestDecision on their own, and all agents can do so at completely different code-defined times, there is no...
    Post by: awjuliani, Mar 11, 2021 in forum: ML-Agents
  10. awjuliani
    Post

    Question OnActionReceived is not called

    Hello, Are you running a training session, or are you simply performing inference with a heuristic method? If it is inference with a heuristic,...
    Post by: awjuliani, Mar 11, 2021 in forum: ML-Agents
  11. awjuliani
    Post

    Memory leak using imitation learning

    Having a large buffer with SAC is not an issue, so long as your machine has the available RAM.
    Post by: awjuliani, Mar 11, 2021 in forum: ML-Agents
  12. awjuliani
    Post

    Memory leak using imitation learning

    Hello. Can you confirm that you do not see a similar increase in memory usage when running SAC without demonstrations? Because SAC uses a large...
    Post by: awjuliani, Mar 10, 2021 in forum: ML-Agents
  13. awjuliani
    Post

    Connect from python environment to Unity environment?

    Hi @spolg, We do not currently support connecting between agents and environments which are hosted on different machines. If you'd like to do...
    Post by: awjuliani, Mar 10, 2021 in forum: ML-Agents
  14. awjuliani
    Post

    Curriculum Learning Logic

    Hi. You are right that by default the WallJump does not require curriculum and does not use it either. We do provide a curriculum yaml, as you...
    Post by: awjuliani, Feb 9, 2021 in forum: ML-Agents
  15. awjuliani
    Post

    MLAgents training progress suddenly declines completely

    I am glad you have been able to arrive at a model which can solve the task. The behavior of the model breaking during training is still a strange...
    Post by: awjuliani, Feb 9, 2021 in forum: ML-Agents
  16. awjuliani
    Post

    Question Train ML Agent on different levels

    Hello. You can run resume to continue training a model on a new environment, but I would recommend having many different agents learning on...
    Post by: awjuliani, Feb 9, 2021 in forum: ML-Agents
  17. awjuliani
    Post

    MLAgents training progress suddenly declines completely

    Thank you for trying these experiments. I see that you also used a modified version of the RollerBall config, which is unfortunately not very...
    Post by: awjuliani, Feb 8, 2021 in forum: ML-Agents
  18. awjuliani
    Post

    Stacked Vectors initial values

    Hello. The stacked vectors are initialized to zeros.
    Post by: awjuliani, Feb 8, 2021 in forum: ML-Agents
  19. awjuliani
    Post

    3D Grid Sensor, Update for ML-Agents 2.1

    Hi mbaske. This seems like a very cool extension of the GridSensor. Thanks for sharing it!
    Post by: awjuliani, Feb 8, 2021 in forum: ML-Agents
  20. awjuliani
    Post

    Is there doc for the yaml files?

    Hello. Here we have a documentation page which explains the configuration file parameters:...
    Post by: awjuliani, Feb 8, 2021 in forum: ML-Agents
Showing results 1 to 20 of 69 Next >
Search Again
Page 1 of 4