From add21c7019b9f8e1961d9543a73db64564689cd7 Mon Sep 17 00:00:00 2001 From: "ys1087@partner.kit.edu" Date: Wed, 27 Aug 2025 12:03:41 +0200 Subject: [PATCH] Clarify that installation must run on GPU node Remove manual installation instructions as PyTorch CUDA dependencies require GPU node --- README.md | 9 +++------ 1 file changed, 3 insertions(+), 6 deletions(-) diff --git a/README.md b/README.md index dd5468d..0225c1b 100644 --- a/README.md +++ b/README.md @@ -64,14 +64,11 @@ The DPPO repository has been adapted to run on the HoReKa cluster. The original 3. **Install the package and dependencies:** ```bash - # Use the provided installation script + # Submit installation job (runs on dev node with GPU) sbatch install_dppo.sh - - # Or install manually: - pip install --upgrade pip - pip install -e . - pip install -e .[gym] # For Gym environments ``` + + Note: Installation must run on a GPU node due to PyTorch CUDA dependencies. The installation script automatically requests appropriate resources. ### Running on HoReKa