Search Unity

Search Results

Search Again
Page 1 of 7
  1. ruoping_unity
    Post

    Advice on Penalising Shots to avoid excessive shots fired ?

    In your decision requester, did you turn on `TakeActionsBetweenDecisions`? That might be one reason that that you're seeing 5 shots when request...
    Post by: ruoping_unity, Sep 13, 2021 in forum: ML-Agents
  2. ruoping_unity
    Post

    Resolved When Training, mlagents displays 0

    Can you elaborate on your steps for launching the experiments? What ML-Agents C# package version are you using? Have you pressed play and see the...
    Post by: ruoping_unity, Sep 13, 2021 in forum: ML-Agents
  3. ruoping_unity
    Post

    Question Unable to reach perfection in extraordinarily simple environment

    Random.Range() seems to be inclusive on min and max value so your input might actually be [0, 1, 2] instead of [0, 1]
    Post by: ruoping_unity, Sep 8, 2021 in forum: ML-Agents
  4. ruoping_unity
    Post

    ML agents: Match-3 Extension doubt

    Our match-3 integration provides a convenient interface for setting up the swap action and the board, so for games that follows the similar...
    Post by: ruoping_unity, Sep 8, 2021 in forum: ML-Agents
  5. ruoping_unity
    Post

    Which parameter affects the RAM usage for SAC?

    If you're running out of memory during training you should consider using smaller batch size. Also the network architecture and network size also...
    Post by: ruoping_unity, Sep 8, 2021 in forum: ML-Agents
  6. ruoping_unity
    Post

    Receive images from camera sensor in python and use custom neural network to take action

    To build your own custom networks, you can use our low-level python API. The DecisionSteps and TerminalSteps you get from each env.step() will...
    Post by: ruoping_unity, Sep 8, 2021 in forum: ML-Agents
  7. ruoping_unity
    Post

    Cannot find Academy

    Hi, what version of ml-agents package you're using? If you're using 1.x or 2.0 version, there's no academy tab and no need to attach academy to an...
    Post by: ruoping_unity, Sep 8, 2021 in forum: ML-Agents
  8. ruoping_unity
    Post

    ML-Agents Package Settings - available in Player?

    Hi @JezMK , Thanks for raising this, we didn't realize this would be confusing to users. I think the confusion comes from the fact that we allow...
    Post by: ruoping_unity, Aug 16, 2021 in forum: ML-Agents
  9. ruoping_unity
    Post

    Cannot use ml-agents in my own game, missing assembly?

    If you're using assembly definition (.asmdef file), you need to add mlagents there. The error message is saying it's missing the assembly to...
    Post by: ruoping_unity, Aug 16, 2021 in forum: ML-Agents
  10. ruoping_unity
    Post

    SAC freeze when working with demonstration

    Behavioral cloning will introduce an extra update which contribute to longer training time, but usually it's not that long. I saw that in your...
    Post by: ruoping_unity, Aug 3, 2021 in forum: ML-Agents
  11. ruoping_unity
    Post

    On the use of actionMask with python API

    Can you reproduce this error with any of our example environments? I verified with our GridWorld environment and the results looks normal with...
    Post by: ruoping_unity, Aug 2, 2021 in forum: ML-Agents
  12. ruoping_unity
    Post

    Question Is the paper about MA-POCA available now?

    Hey we're still working on it and hopefully to have it ready soon. Stay tuned!
    Post by: ruoping_unity, Aug 2, 2021 in forum: ML-Agents
  13. ruoping_unity
    Post

    Feature Request Option to disable rotate with player on Raycast sensors

    This should be fairly easy to implement. I've added this feature request to our lists and hopefully we can get it out soon.
    Post by: ruoping_unity, Aug 2, 2021 in forum: ML-Agents
  14. ruoping_unity
    Post

    Feature Request Is it possible to start curriculum learning from a specified lesson number?

    Feature request received. It's not currently supported but we're happy to add it to our future tasks. Is there a specific use case you can share...
    Post by: ruoping_unity, Jul 29, 2021 in forum: ML-Agents
  15. ruoping_unity
    Post

    Is it possible to take more than one action per timestep?

    Hey correct me if my understanding is wrong since I don't think I fully understand your question. So you can take multiple actions in each...
    Post by: ruoping_unity, Jul 29, 2021 in forum: ML-Agents
  16. ruoping_unity
    Post

    Question time_horizon configuration settings for my case

    time_horizon affects the value estimate (how you calculate expected total reward in the future) in training, more detailed explanation is here. It...
    Post by: ruoping_unity, Jul 28, 2021 in forum: ML-Agents
  17. ruoping_unity
    Post

    Question Activating gravity in OnEpisodeBegin function

    Since we don't have access to your project it's a bit hard to debug your custom environment. But in terms of physics and gravity, ML-Agents...
    Post by: ruoping_unity, Jul 28, 2021 in forum: ML-Agents
  18. ruoping_unity
    Post

    Unity ML-Agents Python Low Level API

    It seems like it's having problem connecting to the Unity build. The first thing I'd suggest is to make sure the mlagents python version in your...
    Post by: ruoping_unity, Jul 28, 2021 in forum: ML-Agents
  19. ruoping_unity
    Post

    Increasing the bandwith of Unity to ML-agent communication

    The best thing I can tell is to use compressed observations with your camera sensors. Also since your observation size is pretty large so it's...
    Post by: ruoping_unity, Jul 28, 2021 in forum: ML-Agents
  20. ruoping_unity
    Post

    Unity ml-agents OnEpisodeBegin() error when multiple agents utilised

    Hi @olmez49! The error you're seeing is because the environment reset is being called multiple times and agents are trying to destroy the...
    Post by: ruoping_unity, Jul 13, 2021 in forum: ML-Agents
Showing results 1 to 20 of 134 Next >
Search Again
Page 1 of 7