3. Train Models on GCP - Pommerman/pytorch-pommerman-rl GitHub Wiki
Run the training script
cd pytorch-pommerman-rl
python main.py --use-gae --env-name PommeFFACompetitionFast-v0 --no-norm --seed 42 --algo a2c --lr-schedule 25000000 --no-vis
Monitor CPU and GPU
### cpu
htop
### gpu utilization
nvidia-smi -l 8000 > gpu_perf.log
CHECK YOUR BILLING WHEN YOU ARE TRAINING YOUR MODEL, STOP IT IF YOU ARE OUT OF CREDITS
Cuda
If you want to see gpu utilization you will likely need cuda. This is specifically if you have an NVIDIA GPU and for Ubuntu.
wget https://developer.download.nvidia.com/compute/cuda/repos/ubuntu1804/x86_64/cuda-ubuntu1804.pin
sudo mv cuda-ubuntu1804.pin /etc/apt/preferences.d/cuda-repository-pin-600
sudo apt-key adv --fetch-keys https://developer.download.nvidia.com/compute/cuda/repos/ubuntu1804/x86_64/7fa2af80.pub
sudo add-apt-repository "deb http://developer.download.nvidia.com/compute/cuda/repos/ubuntu1804/x86_64/ /"
sudo apt-get update
sudo apt-get -y install cuda