How to use Machine Learning AI in Unity! (ML-Agents)

44 thoughts on “How to use Machine Learning AI in Unity! (ML-Agents)”

1. was I supposed to clone repository first?

2. it says cant build wheel for numpy when i try to install mlagents. whats with that?

3. ờ mây zing gút zóp.

4. It's a very good video but can't make it work , we don't have the same parameters anymore I can't whose between discrete or continuous

5. Hmmm…. I followed each and every step (ofcouse with a newer version as of right now) and ran the script.
   The console in Unity gives me constantly this message:

   Fewer observations (0) made than vector observation size (6). The observations will be padded.

   What am I doing wrong and how do I solve that?

6. When running mlagents-learn –help I get an traceback error. I think something is out of date

7. One of the most ridiculously annoying voices I've ever heard on Youtube. Consider getting someone else to do the voiceovers or get speech therapy

8. Traceback (most recent call last):
   
   File "<frozen runpy>", line 198, in _run_module_as_main
   
   File "<frozen runpy>", line 88, in _run_code
   
   File "E:OfficeAim_ball_shootvenvScriptsmlagents-learn.exe__main__.py", line 4, in <module>
   
   File "E:OfficeAim_ball_shootvenvLibsite-packagesmlagentstrainerslearn.py", line 2, in <module>
   
   from mlagents import torch_utils
   
   File "E:OfficeAim_ball_shootvenvLibsite-packagesmlagentstorch_utils__init__.py", line 1, in <module>
   
   from mlagents.torch_utils.torch import torch as torch # noqa
   
   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
   
   File "E:OfficeAim_ball_shootvenvLibsite-packagesmlagentstorch_utilstorch.py", line 6, in <module>
   
   from mlagents.trainers.settings import TorchSettings
   
   File "E:OfficeAim_ball_shootvenvLibsite-packagesmlagentstrainerssettings.py", line 25, in <module>
   
   from mlagents.trainers.cli_utils import StoreConfigFile, DetectDefault, parser
   
   File "E:OfficeAim_ball_shootvenvLibsite-packagesmlagentstrainerscli_utils.py", line 5, in <module>
   
   from mlagents_envs.environment import UnityEnvironment
   
   File "E:OfficeAim_ball_shootvenvLibsite-packagesmlagents_envsenvironment.py", line 12, in <module>
   
   from mlagents_envs.side_channel.side_channel import SideChannel
   
   File "E:OfficeAim_ball_shootvenvLibsite-packagesmlagents_envsside_channel__init__.py", line 5, in <module>
   
   from mlagents_envs.side_channel.default_training_analytics_side_channel import ( # noqa
   
   File "E:OfficeAim_ball_shootvenvLibsite-packagesmlagents_envsside_channeldefault_training_analytics_side_channel.py", line 7, in <module>
   
   from mlagents_envs.communicator_objects.training_analytics_pb2 import (
   
   File "E:OfficeAim_ball_shootvenvLibsite-packagesmlagents_envscommunicator_objectstraining_analytics_pb2.py", line 35, in <module>
   
   _descriptor.FieldDescriptor(
   
   File "E:OfficeAim_ball_shootvenvLibsite-packagesgoogleprotobufdescriptor.py", line 561, in _new_
   
   _message.Message._CheckCalledFromGeneratedFile()
   
   TypeError: Descriptors cannot not be created directly.
   
   If this call came from a _pb2.py file, your generated code is out of date and must be regenerated with protoc >= 3.19.0.If you cannot immediately regenerate your protos, some other possible workarounds are:
   
   1. Downgrade the protobuf package to 3.20.x or lower.
   
   2. Set PROTOCOL_BUFFERS_PYTHON_IMPLEMENTATION=python (but this will use pure-Python parsing and will be much slower).

   More information: https://developers.google.com/protocol-buffers/docs/news/2022-05-06#python-updates

   this is the error i am getting plz help me

9. Thank you for the great video. After dragging and attaching the MoveToGoalAgent script to the Agent, my inspector got two MoveToGoalAgent(script) instead of one behavior parameter and one MoveToGoalAgent(script). I would appreciate it if you gave me a hint to solve that problem.

10. Thanks for a greate tutorial! I've been playing with DL and RL for the last year or so (though not very intesively), and it's one of my main objects of interest, especially RL. And it's a greate pleasure to me to finally discover such things as ML-Agents in general, and Unity integration in particular (why didn't I find it before?). Can't even express now what a huge window of possibilites I see now for myself. A huge field of experiments I wanted to do for a long time. And what an awesome way to create my own environments! I just saw that I can even simply just create the envs in Unity, but make the rest of learning separately in Python!

11. Fantastic tutorial, thank you for all the indepth steps.

    Quick question as you mentioned not bothering with CUDA if you don't have an NVIDIA graphics card and it'll just utilize the CPU instead.

    For those of use with AMD cards, do you know if this toolkit integrates with ROCm or OpenCL?

12. I'm stuck at the part where you need to go to the definition of the agent class, vs just doesn't find it

13. Hi Code monkey , what are the alternatives to making an ai , i am trying to make an ai for a card game like gin rummy and will unity machine learning ai give same actions under same moves?

14. update the trackcheckpoint script. it doesn't have the carcheckpointeventargs, resetcheckpoint and getnextcheckpoint

15. can i do that using a flash drive? and how?

16. I ran into a Traceback error when running "mlagents-learn" @17:40. ('str' object has no attribute '_key') I was able to fix this by installing an older version of the importlib-metadata library: pip install importlib-metadata==4.4

17. OK! 2023, for release 20 of ML agent! the Torch install is now as follows

    pip install torch==1.8.0 torchvision==0.9.0 torchaudio==0.8.0

18. Interesting…… so it's built in to Unity huh…

19. I tried this but I failed at multiple things. First of all, I couldnt find the unity folder right in C: and I ended up using the one in the appdataroaming folder even though there werent any project files in there. Second of all, I placed the virtual environment right in the Unity folder. I was running command prompt as administrator aswell. And i couldnt download the specific version of pytorch. It ended up that i was in my python project without MLagents installed. Just like always

20. It didn't work for me 🙁

21. When I started fallowing this tutorial I would get a lot of errors, but I managed to fix those errors, and then I decided to make my own ML-Agents video. It is vary similar to this video but it shows how I fixed all the errors I was getting. For those that want to fallow my tutorial here is a link: https://youtu.be/RANRz9oyzko. But I just wanted to say thank you code monkey for this awesome video. without it I would not have been able to make my video.

22. Would love another tutorial to Machine Learning in unity.

23. After a few days of following your tutorial little by little (sorry I'm still a newbie),, finally I can understand what you teach, even though there are many changes to the ML-Agents.. but thank you so much Mr.hugo for the knowledge this is very useful for me curious about machine learning.

24. if it doesn´t work on 2023, add a rigid body to the blue cube

25. I know this video is from 2years ago but pls help me. When I want to write the "OnCollisionEnter" ( 24:40), I dont get the suggestion and it doesnt work :/. My game is in 2d can this be the problem??

26. Nothing is working and i cant find a different video tutorial that is actually up to date

27. its back to watching other people achieve my dream

28. dosent work as usual

29. and even if it did work it wont for long

30. hi codemonkey, thanks for the video, but i need help. i stuck on 6:45, i got an error when i ran mlagents-learn –help. it said
    "TypeError: Invalid first argument to `register()`. typing.Dict[mlagents.trainers.settings.RewardSignalType, mlagents.trainers.settings.RewardSignalSettings] is not a class or union type." if you or anyone else know the solution,the ha help is much appreciated

31. I'm too dumb for this I tried to follow it all the way but I can't deal with all those errors and hickups constantly, I'll just wait for ML sandboxes to get good GUIs cause all this Python/pip/C#/Unity making my brain hurt

32. which version of python is being used here?

33. please make a video on "How to implement multiagent from scratch?" with mlagents toolkit.

34. Wow thanks for this

35. good video

36. Really nice video ! I have an idee : The same scenario with an other Agent (agent2). If Agent2 touch agent1, he win

37. For anyone having problems with trying to run `mlagents-learn`, it helps quite significantly if you be VERY mindful that you install the exact same versions.

38. Bro I litterally cant complete it because the docs for installation are not working

39. i have a problem mlagents-learn dont go

40. Hello, i am not getting float values in console when using the ContinuousActions[0]. Just getting Zero every time please help me.

41. Great video, thank you Mr. Code Monkey

42. hello! nice tutorial as always but I have a question. It is possible to continue to train the agent in a build environment? I want use it in my game to the players have a personalized experience, so the agent should learn how each player plays the game. tks in advance

43. Hi from 2023! For whom ever had a problem with pip packages and instaling phyton here is what worked for me lately:
    Unity: 2022.3.10
    Python 3.9.13
    And in venv activated i wrote successively:
    python -m pip install –upgrade pip
    pip install mlagents
    pip3 install torch torchvision torchaudio
    pip install protobuf==3.20.3

    Works for me, cheers!

44. Great video! Im wondering what extensions you use for vscode to have things like autocomplete

Comments are closed.