-
Hi guys,
We've created a model in Python, it is a .pb file.
Could you please advise on how we can convert this to a model file that can be used by Unity Inference?
Also, how far along is Unit…
-
Hi everyone, I work a little with mlagents for make a game where an agent touch one cube, if touch the correct cube done() and reward +1, but if touch the incorrect cube (there is mor than one cube) I…
-
I've followed all Windows installation steps carefully, [here](https://github.com/Unity-Technologies/ml-agents/blob/master/docs/Installation-Windows.md) at the end everything worked on CPU. Then I tri…
-
The version of my mlagents is 0.10.1 and Ubuntu18.04
I build the game and test other algorithms written by myself to train (like TD3)
At the very beginning, everything is OK and the training speed i…
-
Hello guys I made this video showing the project that I made in Unity and trained with ML-Agents.
Hope you will enjoy.
[Matt's youtube video](https://www.youtube.com/watch?v=50oMJ1dNB_o&feature=yout…
-
I have created a custom environment using **MLAgents 0.91**.
I create an instance of the environment via python like this:
`env = UnityEnvironment(file_name=ENV_NAME, worker_id=worker_id)` It works,…
-
When using gym_unity to interact with a built Unity environment, the reward obtained in the `env.step(action)` at the last timestep (when `done = True`) seems to be 0.
Tried this on the Basic and 3D…
-
Hi,
Something wrong had happened when i was test the RollerBall with tensorflow-gpu in windows. I have tried `pip uninstall tensorflow-gpu==2.0` and `pip install tensorflow-gpu==2.0` several times, b…
-
**Describe the bug**
When running the `Colab-UnityEnvironment-2-Train.ipynb` colab notebook the step that runs the Unity environment hangs
from: https://github.com/Unity-Technologies/ml-agents/tre…
-
I am doing a hard test on the model I have trained. So I am exploring variations on the environment to verify the efficiency of my agent in untrained situations.
The problem is that I am running th…