0

I have successfully set up my environment with the necessary installations and confirmed that both PyTorch and TensorFlow have access to CUDA. However, when I train my model using Stable Baselines 3 and Gym, it only utilizes around 10% of my CPU, 50% of CUDA, 1% of the GPU (RTX 3080). I have tried various modifications to the code, but the resource usage remains unchanged. Consequently, I am concerned that there might be an issue with my setup.

I have come across several suggestions that the model might be too small to make full use of the available resources. To test my setup and ensure that my resource utilization is not an issue, I am looking for a sample reinforcement learning code that uses Stable Baselines 3 and Gym, which will fully utilize the power of my RTX 3080 GPU.

If anyone can provide a sample code that maximizes my GPU usage, it would help me to continue my reinforcement learning journey with confidence, knowing that my setup is functioning correctly.

Any assistance or guidance on this matter would be greatly appreciated.

I am a beginner so I were only able to change timesteps and rl methods, I have tried A2C and ppo

Akbar Ali
  • 1
  • 1
  • increase the complexity of your network, e.g., the amount of hidden layers and the amount of neurons in each one and also increase the amount of timesteps. Take a look at https://stable-baselines3.readthedocs.io/en/master/guide/custom_policy.html – gehirndienst Apr 05 '23 at 07:34

0 Answers0