Search Unity

Search Results

Search Again
Page 1 of 2
  1. BotAcademy
    Post

    Saved Replay Buffer 13 MB -> 4KB after Training

    No problems because I'm not using the saved buffer (just tested it for understanding). But if I'd use it I would expect a saved buffer at end of...
    Post by: BotAcademy, Aug 6, 2020 in forum: ML-Agents
  2. BotAcademy
    Post

    [Bot Academy] ML-Agents Playlist on YouTube

    push (new video about the ML-Agents Hyperparameters)
    Post by: BotAcademy, Aug 4, 2020 in forum: ML-Agents
  3. BotAcademy
    Post

    Saved Replay Buffer 13 MB -> 4KB after Training

    When interrupting training, it works fine. Just experienced it when I finished a training run, without interruption. I'd still expect that the...
    Post by: BotAcademy, Aug 4, 2020 in forum: ML-Agents
  4. BotAcademy
    Thread

    Saved Replay Buffer 13 MB -> 4KB after Training

    Hey, just trained the PushBlock example environment with 'save_replay_buffer' set to 'true'. During training the 'last_replay_buffer.hdf5'...
    Thread by: BotAcademy, Aug 1, 2020, 6 replies, in forum: ML-Agents
  5. BotAcademy
    Post

    Buffer Size Parameter - Clarification

    https://github.com/Unity-Technologies/ml-agents/pull/4252
    Post by: BotAcademy, Jul 20, 2020 in forum: ML-Agents
  6. BotAcademy
    Post

    Buffer Size Parameter - Clarification

    Okay :) I can make a PR - will post the PR link here during the next hour
    Post by: BotAcademy, Jul 20, 2020 in forum: ML-Agents
  7. BotAcademy
    Thread

    Buffer Size Parameter - Clarification

    Hey! I was just wondering if I understood the buffer_size parameter correctly. The documentation confused me somehow. Documentation (default =...
    Thread by: BotAcademy, Jul 18, 2020, 4 replies, in forum: ML-Agents
  8. BotAcademy
    Post

    [Bot Academy] ML-Agents Playlist on YouTube

    push (next video about the Hyperparameters for PPO and SAC likely by end of July)
    Post by: BotAcademy, Jul 18, 2020 in forum: ML-Agents
  9. BotAcademy
    Post

    Agents with dynamic Vector Action?

    From what I know, the size of the observation space has to be of a fixed size. I think it is an algorithmic limitation for both PPO and SAC.
    Post by: BotAcademy, Jul 18, 2020 in forum: ML-Agents
  10. BotAcademy
    Post

    Question Cummulative Reward goes down

    does it help setting it to constant?
    Post by: BotAcademy, Jul 15, 2020 in forum: ML-Agents
  11. BotAcademy
    Post

    Question Cummulative Reward goes down

    Maybe it's because of the lack of exploration at the end of training with an environment that still changes to unknown shapes? Are you using PPO...
    Post by: BotAcademy, Jul 14, 2020 in forum: ML-Agents
  12. BotAcademy
    Post

    Question Cummulative Reward goes down

    Entropy looks good but from my understanding I would expect the reward to increase when the entropy goes down. Is there any randomness in your...
    Post by: BotAcademy, Jul 12, 2020 in forum: ML-Agents
  13. BotAcademy
    Post

    Resolved How do I train for more steps?

    Also make sure that the name you set in the Behavior Parameters -> Behavior Name is the same that you defined in the config. Otherwise it will...
    Post by: BotAcademy, Jul 12, 2020 in forum: ML-Agents
  14. BotAcademy
    Post

    Question Weird change in cumulative reward

    Thats interesting. Hopefully someone from the dev team can help you out!
    Post by: BotAcademy, Jul 8, 2020 in forum: ML-Agents
  15. BotAcademy
    Post

    Question Weird change in cumulative reward

    do you have set a max_step for an episode or is it a continuous environment which does not reset? I guess the agent encountered a situation that...
    Post by: BotAcademy, Jul 7, 2020 in forum: ML-Agents
  16. BotAcademy
    Post

    Beta

    I am quite sure that the agent needs to explore more with the complex observations because there are more possibilities, so it takes longer to...
    Post by: BotAcademy, Jul 7, 2020 in forum: ML-Agents
  17. BotAcademy
    Post

    Question training stops with RuntimeError: dictionary changed size during iteration

    I had the same issue. Used Release 3 and the default 3DBall environment with SAC. Only changed the max_steps to 1 million. Occurred at around 700k...
    Post by: BotAcademy, Jul 7, 2020 in forum: ML-Agents
  18. BotAcademy
    Post

    [Bot Academy] ML-Agents Playlist on YouTube

    added a new video about Tensorboard with ML-Agents
    Post by: BotAcademy, Jul 7, 2020 in forum: ML-Agents
  19. BotAcademy
    Post

    Question "The associated script can not be loaded". Can't drag 3DBall.nn model into Behavior Parameters (Scri

    Good to know. Glad that it's working for you now! Enjoy ML-Agents :)
    Post by: BotAcademy, Jul 2, 2020 in forum: ML-Agents
  20. BotAcademy
    Post

    Question "The associated script can not be loaded". Can't drag 3DBall.nn model into Behavior Parameters (Scri

    That goes beyond my knowledge to be honest, so I hope someone else can help you. Chances are quite big that it is not working cause of missing...
    Post by: BotAcademy, Jul 1, 2020 in forum: ML-Agents
Showing results 1 to 20 of 32 Next >
Search Again
Page 1 of 2