Search Unity

  1. Welcome to the Unity Forums! Please take the time to read our Code of Conduct to familiarize yourself with the forum rules and how to post constructively.
  2. Join us on Thursday, June 8, for a Q&A with Unity's Content Pipeline group here on the forum, and on the Unity Discord, and discuss topics around Content Build, Import Workflows, Asset Database, and Addressables!
    Dismiss Notice
New Forum User Notice Content Pipeline Dev Blitz Day

Issues with running example scenes

Discussion in 'ML-Agents' started by jasonrdunne, Mar 3, 2020.

  1. jasonrdunne

    jasonrdunne

    Joined:
    Mar 24, 2019
    Posts:
    4
    I'm having problems getting the 3d ball example to run. I use the command "mlagents-learn C:\Users\jason\OneDrive\Documents\GitHub\ml-agents\config/trainer_config.yaml --run-id=firstRun --train" to begin training. The unity logo appears, and it will begin listening to unity, but after ~40 seconds it will interrupt and fail to save to a directory. Additionally, it doesn't give me any of the stats when training begins, like --debug, --env, etc.
    tl;dr of post: I get the issue WARNING:mlagents.trainers:Unable to save to ./summaries/ppo_timers.json. Make sure the directory exists
    Here is the cmd:


    C:\Users\jason>mlagents-learn C:\Users\jason\OneDrive\Documents\GitHub\ml-agents\config/trainer_config.yaml --run-id=firstRun --train

    WARNING:tensorflow:From c:\python\lib\site-packages\tensorflow_core\python\compat\v2_compat.py:65: disable_resource_variables (from tensorflow.python.ops.variable_scope) is deprecated and will be removed in a future version.

    Instructions for updating:

    non-resource variables are not supported in the long term

    *****(((((MASSIVE UNITY LOGO)))))*****

    Version information:

    ml-agents: 0.15.0.dev0,

    ml-agents-envs: 0.15.0.dev0,

    Communicator API: API-15-dev0,

    TensorFlow: 2.0.1

    WARNING:tensorflow:From c:\python\lib\site-packages\tensorflow_core\python\compat\v2_compat.py:65: disable_resource_variables (from tensorflow.python.ops.variable_scope) is deprecated and will be removed in a future version.

    Instructions for updating:

    non-resource variables are not supported in the long term

    INFO:mlagents_envs:Listening on port 5004. Start training by pressing the Play button in the Unity Editor.
    ((((I PRESS PLAY)))))
    INFO:mlagents.trainers:UnityEnvironment worker 0: environment stopping.

    INFO:mlagents.trainers:Learning was interrupted. Please wait while the graph is generated.

    INFO:mlagents.trainers:Saved Model

    WARNING:mlagents.trainers:Unable to save to ./summaries/firstRun_timers.json. Make sure the directory exists





    If someone could help me troubleshoot this, it would be hugely appreciated.

    Additional info:

    Python version = 3.6

    I cloned the project from github
     
    Last edited: Mar 3, 2020
    jasonrdunne, Mar 3, 2020
    #1

  2. jasonrdunne

    jasonrdunne

    Joined:
    Mar 24, 2019
    Posts:
    4
    I need to present this to my class tomorrow lmao. H E L P
     
    jasonrdunne, Mar 3, 2020
    #2

  3. fariazz

    fariazz

    Joined:
    Nov 21, 2016
    Posts:
    55
    Not sure if this helps, but it might be good to try it out outside of OneDrive
     
    fariazz, Mar 9, 2020
    #3
    ajedir likes this.

  4. rubyreubs

    rubyreubs

    Joined:
    Feb 12, 2019
    Posts:
    1
    This may be because you have your Behaviour Type set to "Heuristic Only" in the Behaviour Parameters attached to your agent in the editor. Hope this helps ;)
     
    rubyreubs, Apr 1, 2020
    #4
    crazylazylife and alsaeed like this.

  5. Xinzz

    Xinzz

    Joined:
    Jun 28, 2015
    Posts:
    67
    same here, any solutions?
     
    Xinzz, Apr 1, 2020
    #5

  6. alsaeed

    alsaeed

    Joined:
    Nov 23, 2015
    Posts:
    1
    Many thanks. god bless
     
    alsaeed, Apr 18, 2020
    #6