Search Unity

Search Results

Search Again
Page 1 of 2
  1. ShirelJosef
    Thread

    No learning when switching from release <10 to release 18

    Hi, I am using PPO to solve an environment using release 4. I can switch to release 10 with no issues and the learning process is the same....
    Thread by: ShirelJosef, Oct 10, 2021, 0 replies, in forum: ML-Agents
  2. ShirelJosef
    Thread

    Which Ml-Agents version to choose for maximum stability

    Hi, I am wondering which mlagents version to use. Currently using the ml-agents release 4 which means: com.unity.ml-agents (C#) v1.2.0 ml-agents...
    Thread by: ShirelJosef, Oct 6, 2021, 1 replies, in forum: ML-Agents
  3. ShirelJosef
    Thread

    Using ml-agents on my own environment

    Hi, I wonder if anyone tried to wrap their OWN environment in order to work with unity. Which means they wrote a wrapper using BaseEnv. I have...
    Thread by: ShirelJosef, Aug 12, 2021, 2 replies, in forum: ML-Agents
  4. ShirelJosef
    Thread

    Do the replay experience buffer is cleared when using Curriculum Learning with SAC?

    Hi, As the title asks, does the ENTIRE experience buffer is cleared when using curriculum learning? If not, we can get the same state once with...
    Thread by: ShirelJosef, Aug 12, 2021, 1 replies, in forum: ML-Agents
  5. ShirelJosef
    Thread

    Question VR and unity in ubuntu

    Hi, I am developing using the ubuntu editor. 1. Does any VR headset is supported in unity and ubuntu together? 2. I currently have Varjo headset...
    Thread by: ShirelJosef, May 9, 2021, 1 replies, in forum: VR
  6. ShirelJosef
    Post

    How to get RayPerceptionSensor values

    You can recover the values by using method: Perceive in Class RayPerceptionSensor. You will need the GetRayPerceptionInput() from Class...
    Post by: ShirelJosef, Nov 23, 2020 in forum: ML-Agents
  7. ShirelJosef
    Post

    My training stops for no reason

    You didn't configure the behavior name correctly, I bet that the agent behavior name in the editor is not the same as in the configuration file....
    Post by: ShirelJosef, Nov 22, 2020 in forum: ML-Agents
  8. ShirelJosef
    Post

    Does learning occur if behavior type is set to heuristic

    my current "naive" plan to is to "break" the self-play, by not changing the learning team (keep it on 0 always), and then each n episode, activate...
    Post by: ShirelJosef, Nov 22, 2020 in forum: ML-Agents
  9. ShirelJosef
    Post

    Does learning occur if behavior type is set to heuristic

    I wanted to use unity self-play mechanism to train two agents against each other, let's say two tanks in an empty world. However, it reduced to...
    Post by: ShirelJosef, Nov 22, 2020 in forum: ML-Agents
  10. ShirelJosef
    Post

    My Agent doesn't try all possible Discrete Actions

    @andrewcoh_unity Seem to me that in certain situations without normalizing the network collapse ("converge") really fast to some weird local...
    Post by: ShirelJosef, Nov 21, 2020 in forum: ML-Agents
  11. ShirelJosef
    Post

    My Agent doesn't try all possible Discrete Actions

    I had this weird issue with continuous control and when I changed normalize: false to true, it solved it. Please tell me if that helped
    Post by: ShirelJosef, Nov 20, 2020 in forum: ML-Agents
  12. ShirelJosef
    Post

    Team learning vs Concurrent learning in Multiagent Environment

    According to https://blogs.unity3d.com/2020/02/28/training-intelligent-adversaries-using-self-play-with-ml-agents/ So it is not a multi agent...
    Post by: ShirelJosef, Nov 20, 2020 in forum: ML-Agents
  13. ShirelJosef
    Thread

    Does learning occur if behavior type is set to heuristic

    I just want to make sure that if I have 2 agents in the same scene, both use the same "behavior name" but one behavior type is inference and the...
    Thread by: ShirelJosef, Nov 20, 2020, 5 replies, in forum: ML-Agents
  14. ShirelJosef
    Post

    Question Adding noise to RayPerceptionSensorComponent3D in ml-agents

    From quick digging I think you might need to change the sensor code: ml-agents/com.unity.ml-agents/Runtime/Sensors/RayPerceptionSensor.cs which...
    Post by: ShirelJosef, Nov 19, 2020 in forum: ML-Agents
  15. ShirelJosef
    Thread

    Self-play with custom snapshots or policies

    Hey, Self play use snapshot of past policies to play against. Is there a way to insert my own .nn file or other policy format as one of these...
    Thread by: ShirelJosef, Nov 18, 2020, 0 replies, in forum: ML-Agents
  16. ShirelJosef
    Post

    Curriculum learning and self-play

    Thank you for your reply. I understand the concept of curriculum learning, but my issue is with advancing the lessons. I use a simple +1 -1...
    Post by: ShirelJosef, Oct 19, 2020 in forum: ML-Agents
  17. ShirelJosef
    Thread

    Curriculum learning and self-play

    Hi, 1. Question about mean reward in self-play: maybe a silly question, but is is correct that the reward values that are written to tensor board...
    Thread by: ShirelJosef, Oct 8, 2020, 6 replies, in forum: ML-Agents
  18. ShirelJosef
    Post

    OnActionReceived() order of execution with respect to MonoBehaviour

    @henrypeteet thank you for taking the time to answer my question. As you recommended I am trying to avoid "breaking" the flow, as not knowing the...
    Post by: ShirelJosef, Oct 5, 2020 in forum: ML-Agents
  19. ShirelJosef
    Post

    Forcing PPO to be more explorative

    If you look in the training configuration file documentation then some parameters are associated with exploration, such as beta in PPO...
    Post by: ShirelJosef, Oct 1, 2020 in forum: ML-Agents
  20. ShirelJosef
    Thread

    OnActionReceived() order of execution with respect to MonoBehaviour

    Hi, My agent have a script that manipulate it's movement during FixedUpdate (CatDynamics.cs). I have another script (CatAgent.cs) that inherits...
    Thread by: ShirelJosef, Sep 30, 2020, 10 replies, in forum: ML-Agents
Showing results 1 to 20 of 21 Next >
Search Again
Page 1 of 2