commit
5bd8644025
3 changed files with 4 additions and 19 deletions
19
docs/HPC.md
19
docs/HPC.md
|
|
@ -74,25 +74,6 @@ After installation, run these commands to ensure your environment is set up corr
|
||||||
python -c "import torch, jax; print(f'GPU: {torch.cuda.is_available()}'); print(f'JAX: {jax.devices()}')"
|
python -c "import torch, jax; print(f'GPU: {torch.cuda.is_available()}'); print(f'JAX: {jax.devices()}')"
|
||||||
```
|
```
|
||||||
|
|
||||||
## PR Verification (Quick Start)
|
|
||||||
|
|
||||||
If you are a reviewer verifying a PR, run this single block:
|
|
||||||
|
|
||||||
```bash
|
|
||||||
git checkout <pr-branch>
|
|
||||||
module swap cluster/donphan
|
|
||||||
qsub -I -l nodes=1:gpus=1
|
|
||||||
|
|
||||||
# Inside the interactive session:
|
|
||||||
cd "$PBS_O_WORKDIR"
|
|
||||||
bash scripts/hpc/install.sh
|
|
||||||
python -c "import torch, jax; print(torch.cuda.is_available()); print(jax.devices())"
|
|
||||||
exit
|
|
||||||
|
|
||||||
# Verify batch submission
|
|
||||||
qsub scripts/hpc/train.pbs
|
|
||||||
```
|
|
||||||
|
|
||||||
## Managing Dependencies
|
## Managing Dependencies
|
||||||
|
|
||||||
`env/hpc/requirements.txt` is auto-generated from `pyproject.toml`. To regenerate:
|
`env/hpc/requirements.txt` is auto-generated from `pyproject.toml`. To regenerate:
|
||||||
|
|
|
||||||
|
|
@ -19,6 +19,8 @@ if [ -n "$PBS_O_WORKDIR" ]; then
|
||||||
cd "$PBS_O_WORKDIR"
|
cd "$PBS_O_WORKDIR"
|
||||||
fi
|
fi
|
||||||
|
|
||||||
|
mkdir "${PBS_O_WORKDIR}/runs"
|
||||||
|
|
||||||
# Mirror configs to $VSC_DATA to avoid home quota limits (3GB)
|
# Mirror configs to $VSC_DATA to avoid home quota limits (3GB)
|
||||||
# vsc-venv manages environments relative to the requirements file
|
# vsc-venv manages environments relative to the requirements file
|
||||||
PROJ_NAME=$(basename "$PWD")
|
PROJ_NAME=$(basename "$PWD")
|
||||||
|
|
|
||||||
|
|
@ -6,6 +6,8 @@
|
||||||
#PBS -N brittlestar-ppo
|
#PBS -N brittlestar-ppo
|
||||||
#PBS -l nodes=1:ppn=8
|
#PBS -l nodes=1:ppn=8
|
||||||
#PBS -l walltime=24:00:00
|
#PBS -l walltime=24:00:00
|
||||||
|
#PBS -o runs/brittlestar-ppo.o$PBS_JOBID
|
||||||
|
#PBS -e runs/brittlestar-ppo.e$PBS_JOBID
|
||||||
|
|
||||||
set -euo pipefail
|
set -euo pipefail
|
||||||
|
|
||||||
|
|
|
||||||
Reference in a new issue