mingfeisun/deepmimic_mujoco

DeepMimic Mujoco version; Python only; No Bullet, No C++

deepmimic_mujoco's Issues

pip dependencies required but not listed

Issue

this project required many python dependencies to be installed using pip, but does not list which ones need to be installed.

Known required and missing

gym
tensorflow
joblib
mujoco_py Instructions: https://github.com/openai/mujoco-py#install-mujoco

Problem running a demo on Windows 10 using conda

I've tried to run a DeepMimic_mujoco demo on Windows 10 using conda with python 3.6
Here is what happens when I try to run env _torque_test.py

(base) C:\Users\Utilisateur\Desktop\DeepMimic_mujoco-master\src>py env_torque_test.py
running build_ext
building 'mujoco_py.cymj' extension
Traceback (most recent call last):
File "env_torque_test.py", line 2, in
from dp_env_v3 import DPEnv
File "C:\Users\Utilisateur\Desktop\DeepMimic_mujoco-master\src\dp_env_v3.py", line 10, in
from mujoco_py import load_model_from_xml, MjSim, MjViewer
File "c:\users\utilisateur\desktop\mujoco-py-1.50.1.0\mujoco_py_init_.py", line 1, in
from mujoco_py.builder import cymj, ignore_mujoco_warnings, functions, MujocoException
File "c:\users\utilisateur\desktop\mujoco-py-1.50.1.0\mujoco_py\builder.py", line 283, in
cymj = load_cython_ext(mjpro_path)
File "c:\users\utilisateur\desktop\mujoco-py-1.50.1.0\mujoco_py\builder.py", line 54, in load_cython_ext
cext_so_path = builder.build()
File "c:\users\utilisateur\desktop\mujoco-py-1.50.1.0\mujoco_py\builder.py", line 131, in build
dist.run_commands()
File "C:\Users\Utilisateur\Anaconda3\lib\distutils\dist.py", line 955, in run_commands
self.run_command(cmd)
File "C:\Users\Utilisateur\Anaconda3\lib\distutils\dist.py", line 974, in run_command
cmd_obj.run()
File "C:\Users\Utilisateur\Anaconda3\lib\site-packages\Cython\Distutils\old_build_ext.py", line 186, in run
_build_ext.build_ext.run(self)
File "C:\Users\Utilisateur\Anaconda3\lib\distutils\command\build_ext.py", line 339, in run
self.build_extensions()
File "c:\users\utilisateur\desktop\mujoco-py-1.50.1.0\mujoco_py\builder.py", line 75, in build_extensions
build_ext.build_extensions(self)
File "C:\Users\Utilisateur\Anaconda3\lib\site-packages\Cython\Distutils\old_build_ext.py", line 194, in build_extensions
self.build_extension(ext)
File "C:\Users\Utilisateur\Anaconda3\lib\distutils\command\build_ext.py", line 533, in build_extension
depends=ext.depends)
File "C:\Users\Utilisateur\Anaconda3\lib\distutils_msvccompiler.py", line 304, in compile
self.initialize()
File "C:\Users\Utilisateur\Anaconda3\lib\distutils_msvccompiler.py", line 197, in initialize
vc_env = _get_vc_env(plat_spec)
File "C:\Users\Utilisateur\Anaconda3\lib\distutils_msvccompiler.py", line 85, in _get_vc_env
raise DistutilsPlatformError("Unable to find vcvarsall.bat")
distutils.errors.DistutilsPlatformError: Unable to find vcvarsall.bat

Any help would be great !

How to train the policy for spinkick?

Thanks for your great work, could you please offer some details about how to design the reward function for the task of spinkick?

How do you generate the humanoid3d.txt into .xml files?

As the title tells, I'm curious about this. Since I'm trying to transform the deepmimic character file into urdf format, searching for some methods.

Can I convert your deepmimic mujoco xml to the format that can be loaded into unity3D?

Is there any way that I can convert your deepmimic mujoco xml to the format that can be loaded into unity3D? How should I do that?

wrong policy

I run the code successfuly.However,the agent can not walk ,run after training.Do I miss anything?
Any help would be great !

what's the hyper-params used to train imitations of walking/running?

Thanks for your great work!
I followed #3 to replicate the results but failed. I am wondering what I am doing wrong. Could you post your hyper-params of trpo? I think that will help a lot.

Starting Code for Training

Hello

I have recently installed and started training with this system.
The robot shape and mass was modified in the .xml file to match with the robot in our lab.

For starting the training, in the dp_env_v3.py file, and in the step function:

• the variable "data" was not define, and I assumed it is "self.sim.data"
• should I uncomment the code for self.step_len ? Does this change the simulation step size or something?

I have been trying to train a robot to walk with:
mpirun --allow-run-as-root -np 32 python3 trpo.py (my cpu has 32 threads)
After 1 million time steps, it usually does something to slowly fall forwards. or stand still. Any tips for training?

Have you successfully trained walking or other advanced moves with this set of code?

version issue and about OpenGL

Hello! When I run the dp_env_v3.py, there is a mistake.

mujoco_env.MujocoEnv.__init__(self, xml_file_path, 1)
TypeError: __init__() missing 1 required positional argument: 'observation_space'

My Mujoco is 2.1.0 and the mujoco-py is 2.1.2.14.
Is it a version issue?
Any help would be great !

And I would like to know if OpenGL3.3 needs to be installed on a Linux server to run this code.
Although the server is equipped with RTX 3080 and graphics card driver 510.47.03, my server can still only install OpenGL3.1 Mesa 21.2.6. Is this a problem with graphics card drivers? Is there anything else I need to do to upgrade OpenGL version to 3.2 or higher?