Search Unity

Search Results

Search Again
Page 1 of 3
  1. StewedHarry
    Post

    Bug Window size in Unity executable small after training with a build

    I've checked to see whether this bug persists on another machine. It doesn't. It only occurs on the machine which trained with the executable.
    Post by: StewedHarry, Sep 27, 2020 in forum: ML-Agents
  2. StewedHarry
    Post

    Bug Window size in Unity executable small after training with a build

    I am also unable to maximise the window. This occurs for every build I have made with unity using ML-agents
    Post by: StewedHarry, Sep 27, 2020 in forum: ML-Agents
  3. StewedHarry
    Thread

    Bug Window size in Unity executable small after training with a build

    The window size for all my Unity executables is really small (the same size as when training using executables) after training with multiple...
    Thread by: StewedHarry, Sep 27, 2020, 4 replies, in forum: ML-Agents
  4. StewedHarry
    Post

    Question -num-envs=<n> help - what are Unity instances

    I mean that I have multiple duplicated agents in a single scene. But within this scene multiple environments are training in parallel, like the...
    Post by: StewedHarry, Sep 27, 2020 in forum: ML-Agents
  5. StewedHarry
    Post

    Question -num-envs=<n> help - what are Unity instances

    Sorry I meant to say multiple instances of the same agent in the scene, each contributing to the same policy.
    Post by: StewedHarry, Sep 26, 2020 in forum: ML-Agents
  6. StewedHarry
    Post

    Question -num-envs=<n> help - what are Unity instances

    In what scenarios would the agent be trained differently with multiple Unity instances? I currently have a scene training with multiple agents....
    Post by: StewedHarry, Sep 25, 2020 in forum: ML-Agents
  7. StewedHarry
    Thread

    Question -num-envs=<n> help - what are Unity instances

    I'm looking for ways to speed up training and found this in the documentation: Training Using Concurrent Unity Instances. What is meant here by...
    Thread by: StewedHarry, Sep 24, 2020, 9 replies, in forum: ML-Agents
  8. StewedHarry
    Post

    Question max_step with multiple agents

    Ok great, I think I understand. Thanks for all the info. I wish I could remember where I came across the multiplication of step by the number of...
    Post by: StewedHarry, Sep 8, 2020 in forum: ML-Agents
  9. StewedHarry
    Post

    Question max_step with multiple agents

    I'm trying to train an agent with other agents in the scene running on inference. I remember reading somewhere that either the max step or step...
    Post by: StewedHarry, Sep 8, 2020 in forum: ML-Agents
  10. StewedHarry
    Post

    Question How do I stop my agent moving really fast during inference?

    This worked. Thanks! Timescale was set at 20 during inference for some reason
    Post by: StewedHarry, Sep 7, 2020 in forum: ML-Agents
  11. StewedHarry
    Thread

    Question max_step with multiple agents

    I remember reading something a while ago, either on this forum or in the documentation, that you need to times the max step of each agent by the...
    Thread by: StewedHarry, Sep 7, 2020, 5 replies, in forum: ML-Agents
  12. StewedHarry
    Post

    Running tests consecutively

    You might need to properly setup and teardown each test: [SetUp] public void Init() { // loads a new scene...
    Post by: StewedHarry, Sep 4, 2020 in forum: Testing & Automation
  13. StewedHarry
    Post

    Question How do I stop my agent moving really fast during inference?

    I'm using the same control code for training and inference. I imagine I am experience studder-step, but I didn't anticipate such a large...
    Post by: StewedHarry, Aug 28, 2020 in forum: ML-Agents
  14. StewedHarry
    Post

    Question How do I stop my agent moving really fast during inference?

    This is occurring during inference, which is why I'm confused. I thought inference was supposed to run at normal time. Where do I find Player...
    Post by: StewedHarry, Aug 27, 2020 in forum: ML-Agents
  15. StewedHarry
    Thread

    Question How do I stop my agent moving really fast during inference?

    My agent seems to move extremely fast during inference. I think may have something to do with the frames per second during inference - 60-90fps. I...
    Thread by: StewedHarry, Aug 27, 2020, 6 replies, in forum: ML-Agents
  16. StewedHarry
    Post

    Question Configuring training parameters for maze-like environment.

    I'll try removing the stack and being more patient with LSTM - it seems as though the agents ability to remember where it has been is crucial.
    Post by: StewedHarry, Aug 25, 2020 in forum: ML-Agents
  17. StewedHarry
    Post

    Question Configuring training parameters for maze-like environment.

    There are 20 observations in the stack, 2 for each tile position, so that accounts for 10 previously visited ties. This should be plenty for any...
    Post by: StewedHarry, Aug 25, 2020 in forum: ML-Agents
  18. StewedHarry
    Thread

    Question Configuring training parameters for maze-like environment.

    I'm trying to create an agent which can traverse a maze-like environment, but the resulting behaviour has not lived up to my expectations. I have...
    Thread by: StewedHarry, Aug 25, 2020, 3 replies, in forum: ML-Agents
  19. StewedHarry
    Thread

    Question What happens when end episode is called when one agent is training and another is using inference?

    I'm training two agents in cycles where one will be using inference and the other will be training. If the inferring agent calls EndEpisode while...
    Thread by: StewedHarry, Aug 15, 2020, 1 replies, in forum: ML-Agents
  20. StewedHarry
    Thread

    Question How can I get the Vector3 position of RayPerceptionSensor3D ray-hit?

    I am trying to convert the location of a ray from a RayPerceptionSensor component, however i'm having difficulties. Here is the code: public...
    Thread by: StewedHarry, Aug 11, 2020, 0 replies, in forum: ML-Agents
Showing results 1 to 20 of 45 Next >
Search Again
Page 1 of 3