Search Unity

Search Results

Search Again
Page 1 of 2
  1. kokimitsunami
    Post

    Question Why is the performance of inference so bad?

    Hi @Thorce I believe that selfplay works even when training multiple agents. I've done it myself. You can do it by specifying training settings...
    Post by: kokimitsunami, Jul 28, 2023 in forum: ML-Agents
  2. kokimitsunami
    Post

    ML-Agents/ How can I check the parameter size (or number) of my model?

    Hi @Chika-S I quickly made a python script to check the parameter size of an NN model. I hope this helps. import onnx from onnx import...
    Post by: kokimitsunami, Jul 28, 2023 in forum: ML-Agents
  3. kokimitsunami
    Post

    Question making cars recover from stuck position

    In addition to tweaking reward functions, it might worth considering utilizing action masking. It specifies that some actions are impossible for...
    Post by: kokimitsunami, Jun 21, 2023 in forum: ML-Agents
  4. kokimitsunami
    Post

    Question Reward setting in POCA algorithm

    The both rewards are important to understand agents' behavior, but I would pay attention to the group reward more because it's directly tied to...
    Post by: kokimitsunami, Jun 12, 2023 in forum: ML-Agents
  5. kokimitsunami
    Post

    Question Reward setting in POCA algorithm

    This page says that "A positive group reward indicates the whole group's accomplishments or desired behaviors. Every agent in the group will...
    Post by: kokimitsunami, Jun 8, 2023 in forum: ML-Agents
  6. kokimitsunami
    Post

    Question Ray perception sensor 3D show the result? ml-agents

    Hi, I've posted on how to check Ray Perception Sensor 3D in the below thread:...
    Post by: kokimitsunami, May 22, 2023 in forum: ML-Agents
  7. kokimitsunami
    Post

    Question My Agent Keeps Doing A Discrete Action Even Though I Gave It A Negative Reward

    Hi, I think here you should use AddReward() instead of SetReward(). AddReward() increments reward by the provided value, while SetReward()...
    Post by: kokimitsunami, May 3, 2023 in forum: ML-Agents
  8. kokimitsunami
    Post

    Question Dynamically loading Onnx file and setting as the model for an Agent

    It may depend on the version of ML-Agents and Barracuda you are using, but I was able to update onnx files for ML-Agents dynamically with the...
    Post by: kokimitsunami, Apr 28, 2023 in forum: ML-Agents
  9. kokimitsunami
    Post

    How to create these perception rays?

    Hi, I recently posted on how to check Ray Perception Sensor 3D in the below thread:...
    Post by: kokimitsunami, Apr 24, 2023 in forum: ML-Agents
  10. kokimitsunami
    Post

    Question When training multiple instances of the same agent, do all agents have to be in view?

    I'm not sure about the answer to your question, but there is another way to speed up training. You can use built environment executable rather...
    Post by: kokimitsunami, Apr 13, 2023 in forum: ML-Agents
  11. kokimitsunami
    Post

    Question Limit possible actions & deep action

    Regarding the question 1, yes, there is a way to mask specific actions depending on the game situation. You can refer the page below on how to use...
    Post by: kokimitsunami, Apr 11, 2023 in forum: ML-Agents
  12. kokimitsunami
    Post

    Question How to check Ray perception Sensor 3D?

    You can check the values from Ray Perception Sensor like below: using Unity.MLAgents.Sensors; private void checkRayCast() {...
    Post by: kokimitsunami, Apr 11, 2023 in forum: ML-Agents
  13. kokimitsunami
    Post

    Question How to handle a cooperative multi-agent environment observation ?

    In the github documentation below, there are some descriptions on multi-agent:...
    Post by: kokimitsunami, Apr 11, 2023 in forum: ML-Agents
  14. kokimitsunami
    Post

    Question Masking specific combinations of values in discrete action branches for a board game

    Yes, there is a way to mask specific actions depending on the game situation. You can refer the page below on how to use it:...
    Post by: kokimitsunami, Apr 11, 2023 in forum: ML-Agents
  15. kokimitsunami
    Post

    SAC long train time on 12 cores AMD 5900X

    Hi, I'd like to try to change the number of CPU cores for training as mentioned above. I found where the code is located but cannot find any...
    Post by: kokimitsunami, Nov 30, 2022 in forum: ML-Agents
  16. kokimitsunami
    Post

    Official Post Your ML-Agents Project

    Hi ML Agents community, we developed a playable game demo, in which you can battle against the machine learned knight implemented using ML-Agents....
    Post by: kokimitsunami, Sep 16, 2022 in forum: ML-Agents
  17. kokimitsunami
    Thread

    Question Lots of GCs have occurred during NN model run

    As the Unity Profiler timeline screenshot attached shows, it appears that a lot of GCs is occurring while myagent's NN model is running,...
    Thread by: kokimitsunami, Sep 1, 2022, 1 replies, in forum: ML-Agents
  18. kokimitsunami
    Post

    How do I select/swap networks at run time?

    I've done that for changing difficulty level for my game setting. Switching NN models on runtime can be implemented by using SetModel method as...
    Post by: kokimitsunami, Aug 31, 2022 in forum: ML-Agents
  19. kokimitsunami
    Post

    basic tips to increase training stability

    It totally depends on what task you want to perform, in my case I reduced the size of the neural network and the training stabilized. My neural...
    Post by: kokimitsunami, Jun 24, 2022 in forum: ML-Agents
  20. kokimitsunami
    Post

    Performance when multiple agents exist

    Does anyone have any idea on this? I appreciate if if you could provide any pointers.
    Post by: kokimitsunami, Jun 13, 2022 in forum: ML-Agents
Showing results 1 to 20 of 25 Next >
Search Again
Page 1 of 2