Bug "TypeError: Descriptors cannot be created directly" error when running mlagents-learn

Been following the installation steps https://github.com/Unity-Technologies/ml-agents/blob/release_18_docs/docs/Installation.md and currently I'm getting stuck on the Installing mlagents step, after installation when I try running mlagents-learn --help I see this in the cmd:


Traceback (most recent call last):
File "C:\Users\Jax\AppData\Local\Programs\Python\Python38\lib\runpy.py", line 192, in _run_module_as_main
return _run_code(code, main_globals, None,
File "C:\Users\Jax\AppData\Local\Programs\Python\Python38\lib\runpy.py", line 85, in _run_code
exec(code, run_globals)
File "D:\Users\Jax\Documents\Work\Projects\Machine Learning\python-envs\ml-env\Scripts\mlagents-learn.exe\__main__.py", line 5, in <module>
File "d:\users\jax\documents\work\projects\machine learning\python-envs\ml-env\lib\site-packages\mlagents\trainers\learn.py", line 2, in <module>
from mlagents import torch_utils
File "d:\users\jax\documents\work\projects\machine learning\python-envs\ml-env\lib\site-packages\mlagents\torch_utils\__init__.py", line 1, in <module>
from mlagents.torch_utils.torch import torch as torch # noqa
File "d:\users\jax\documents\work\projects\machine learning\python-envs\ml-env\lib\site-packages\mlagents\torch_utils\torch.py", line 6, in <module>
from mlagents.trainers.settings import TorchSettings
File "d:\users\jax\documents\work\projects\machine learning\python-envs\ml-env\lib\site-packages\mlagents\trainers\settings.py", line 25, in <module>
from mlagents.trainers.cli_utils import StoreConfigFile, DetectDefault, parser
File "d:\users\jax\documents\work\projects\machine learning\python-envs\ml-env\lib\site-packages\mlagents\trainers\cli_utils.py", line 5, in <module>
from mlagents_envs.environment import UnityEnvironment
File "d:\users\jax\documents\work\projects\machine learning\python-envs\ml-env\lib\site-packages\mlagents_envs\environment.py", line 34, in <module>
from mlagents_envs.communicator_objects.command_pb2 import STEP, RESET
File "d:\users\jax\documents\work\projects\machine learning\python-envs\ml-env\lib\site-packages\mlagents_envs\communicator_objects\command_pb2.py", line 32, in <module>
_descriptor.EnumValueDescriptor(
File "d:\users\jax\documents\work\projects\machine learning\python-envs\ml-env\lib\site-packages\google\protobuf\descriptor.py", line 755, 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

 

Did you install into a brand-new python virtual environment, or into an existing python environment?

If into an existing python environment, you might consider creating a new virtual environment, and installing into that. Or downgrading protobuf to 3.20.3.

 

I believe I installed into a brand-new python virtual environment, I have more experience with Unity than with python, I'm not sure how to go about downgrading protobuf. I'm also still seeing the "WARNING: You are using pip version 19.2.3, however version 23.0 is available" message even though I've already ran the upgrade pip command on the vm.

 

- `pip install -U pip` to upgrade pip, I uusllay do `pip install -U pip wheel`, to get both out of the way
- `pip install protobuf==3.20.3` to downgrade protobuf
- are you sure you acitvated your enviornment by the way?
- on linux and mac you can do `which python` and `which pip` to check whether the enviornemtn is activated . I'm not sure how to do that on windows though
- you can do `pip freeze` to list all installed python packages
- to activate your python environment, say you created venv at `venv`. so do `source`, oh wait thats linux. I dont know how to activate windows python enviornments, sorry

 

I'm following the Virtual Environment setup here https://github.com/Unity-Technologi...ase_18_docs/docs/Using-Virtual-Environment.md and I'm pretty certain I'm activating the environment correctly. I ran pip install protobuf==3.20.3 and I'm seeing a different error now when I run mlagents-learn --help


Traceback (most recent call last):
File "C:\Users\Jax\AppData\Local\Programs\Python\Python38\lib\runpy.py", line 192, in _run_module_as_main
return _run_code(code, main_globals, None,
File "C:\Users\Jax\AppData\Local\Programs\Python\Python38\lib\runpy.py", line 85, in _run_code
exec(code, run_globals)
File "D:\Users\Jax\Documents\Work\Projects\Machine Learning\Machine Learning\python-envs\ml-env\Scripts\mlagents-learn.exe\__main__.py", line 4, in <module>
File "D:\Users\Jax\Documents\Work\Projects\Machine Learning\Machine Learning\python-envs\ml-env\lib\site-packages\mlagents\trainers\learn.py", line 13, in <module>
from mlagents.trainers.trainer_controller import TrainerController
File "D:\Users\Jax\Documents\Work\Projects\Machine Learning\Machine Learning\python-envs\ml-env\lib\site-packages\mlagents\trainers\trainer_controller.py", line 13, in <module>
from mlagents.trainers.env_manager import EnvManager, EnvironmentStep
File "D:\Users\Jax\Documents\Work\Projects\Machine Learning\Machine Learning\python-envs\ml-env\lib\site-packages\mlagents\trainers\env_manager.py", line 12, in <module>
from mlagents.trainers.policy import Policy
File "D:\Users\Jax\Documents\Work\Projects\Machine Learning\Machine Learning\python-envs\ml-env\lib\site-packages\mlagents\trainers\policy\__init__.py", line 1, in <module>
from mlagents.trainers.policy.policy import Policy # noqa
File "D:\Users\Jax\Documents\Work\Projects\Machine Learning\Machine Learning\python-envs\ml-env\lib\site-packages\mlagents\trainers\policy\policy.py", line 10, in <module>
from mlagents.trainers.buffer import AgentBuffer
File "D:\Users\Jax\Documents\Work\Projects\Machine Learning\Machine Learning\python-envs\ml-env\lib\site-packages\mlagents\trainers\buffer.py", line 97, in <module>
class AgentBufferField(list):
File "D:\Users\Jax\Documents\Work\Projects\Machine Learning\Machine Learning\python-envs\ml-env\lib\site-packages\mlagents\trainers\buffer.py", line 211, in AgentBufferField
self, pad_value: np.float = 0, dtype: np.dtype = np.float32
File "D:\Users\Jax\Documents\Work\Projects\Machine Learning\Machine Learning\python-envs\ml-env\lib\site-packages\numpy\__init__.py", line 305, in __getattr__
raise AttributeError(__former_attrs__[attr])
AttributeError: module 'numpy' has no attribute 'float'.
`np.float` was a deprecated alias for the builtin `float`. To avoid this error in existing code, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.
The aliases was originally deprecated in NumPy 1.20; for more details and guidance see the original release note at:
https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations

 

Think I figured it out, it was running the wrong version of python even after uninstalling, script is running correctly now.

 

Hello, i have the same problem. can you please tell me how you figured it out with a bit more detail?
while i was trying to install ML Agents i uninstalled my new python version via the installer and then installed the older version (3.9.13 | 64bit).
The CMD currently shows that the python version is 3.9.13 (which i want) & i also have an active virtual environment.

I Typed:

(venv) F:\UnityProjects\MachineLearning>mlagents-learn --help

Error I Got:

Traceback (most recent call last):
File "C:\Users\User\venv\Scripts\mlagents-learn-script.py", line 33, in <module>
sys.exit(load_entry_point('mlagents==0.30.0', 'console_scripts', 'mlagents-learn')())
File "C:\Users\User\venv\Scripts\mlagents-learn-script.py", line 25, in importlib_load_entry_point
return next(matches).load()
File "F:\python old\lib\importlib\metadata.py", line 86, in load
module = import_module(match.group('module'))
File "F:\python old\lib\importlib\__init__.py", line 127, in import_module
return _bootstrap._gcd_import(name[level:], package, level)
File "<frozen importlib._bootstrap>", line 1030, in _gcd_import
File "<frozen importlib._bootstrap>", line 1007, in _find_and_load
File "<frozen importlib._bootstrap>", line 986, in _find_and_load_unlocked
File "<frozen importlib._bootstrap>", line 680, in _load_unlocked
File "<frozen importlib._bootstrap_external>", line 850, in exec_module
File "<frozen importlib._bootstrap>", line 228, in _call_with_frames_removed
File "C:\Users\User\venv\lib\site-packages\mlagents\trainers\learn.py", line 2, in <module>
from mlagents import torch_utils
File "C:\Users\User\venv\lib\site-packages\mlagents\torch_utils\__init__.py", line 1, in <module>
from mlagents.torch_utils.torch import torch as torch # noqa
File "C:\Users\User\venv\lib\site-packages\mlagents\torch_utils\torch.py", line 6, in <module>
from mlagents.trainers.settings import TorchSettings
File "C:\Users\User\venv\lib\site-packages\mlagents\trainers\settings.py", line 25, in <module>
from mlagents.trainers.cli_utils import StoreConfigFile, DetectDefault, parser
File "C:\Users\User\venv\lib\site-packages\mlagents\trainers\cli_utils.py", line 5, in <module>
from mlagents_envs.environment import UnityEnvironment
File "C:\Users\User\venv\lib\site-packages\mlagents_envs\environment.py", line 12, in <module>
from mlagents_envs.side_channel.side_channel import SideChannel
File "C:\Users\User\venv\lib\site-packages\mlagents_envs\side_channel\__init__.py", line 5, in <module>
from mlagents_envs.side_channel.default_training_analytics_side_channel import ( # noqa
File "C:\Users\User\venv\lib\site-packages\mlagents_envs\side_channel\default_training_analytics_side_channel.py", line 7, in <module>
from mlagents_envs.communicator_objects.training_analytics_pb2 import (
File "C:\Users\User\venv\lib\site-packages\mlagents_envs\communicator_objects\training_analytics_pb2.py", line 35, in <module>
_descriptor.FieldDescriptor(
File "C:\Users\User\venv\lib\site-packages\google\protobuf\descriptor.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


I also typed "(venv) F:\UnityProjects\MachineLearning>mlagents-learn". It didn't work either

 

I was only able to get things working after switching the python version to 3.7.0, any other version was having problems.

 

Which version of python were you using? What with what version of python did it work. I am using Python 3.10.11 which doesn't work.

 

it's an error with protobuf ( protocol buffers ) library
just do
pip install protobuf==3.20

 

I training my agent，there is two behavior（forward and rotation). When i start, a warnning is appearing as below:
ml-agents-release_17\ml-agents\mlagents\trainers\torch\utils.py:314: UserWarning: The use of `x.T` on tensors of dimension other than 2 to reverse their shape is deprecated and it will throw an error in a future release. Consider `x.mT` to transpose batches of matricesor `x.permute(*torch.arange(x.ndim - 1, -1, -1))` to reverse the dimensions of a tensor. (Triggered internally at C:\actions-runner\_work\pytorch\pytorch\builder\windows\pytorch\aten\src\ATen\native\TensorShape.cpp:2318.)

And my agent continues rotating and not go forward.

 

ml-agents-release_17 is an old version. Maybe try a more recent one ...