Readiness
- Readiness
- Need >5GB
- Smoke
- not_run
- Artifacts
- not_run
- Production
- need_gt5gb_fampnn_py310_torch_pyg_hydra_sif_pending
- Storage
- 5-20GB
- Strategy
- Python 3.10 PyTorch 2.4.1 CUDA12.1 + PyG/Hydra SIF or ext4-backed overlay; reuse bundled FAMPNN weights.
- Next
- Build runtime on ext4/SIF, run small seq_design.py or pack.py example, verify designed sequences/PDB/CSV artifacts, then wire adapter.
No obvious large model/database dependency identified; validate upstream docs before install. Worker evidence 2026-07-01: source https://github.com/fnachon/fampnn revision 58141f0920b4c178a65de4ef91c1dbe92e177d87 staged at /media/nik/seagate_nik/bio_server/tools/fampnn/source. Tool folder 762MB including source 445MB, weights 114MB, data 159MB, pip_cache 317MB. No Slurm benchmark: runtime deps incomplete. Missing torch, torch_geometric, hydra, omegaconf, torchtyping, timm, natsort, einops, pandas, tqdm; pip installs to exFAT target hung/stalled around Torch/Torch-Geometric. Shared python312_slim.sif lacks required deps. Source imports partly, but inference scripts cannot run. 2026-07-07 16:42 +0800: SOP Production Blocked audit: moved to Need >5GB because only source/weights/data are staged and the real runtime is incomplete; current evidence includes generic-smoke only and missing Torch/Torch-Geometric/Hydra stack. 2026-07-17T01:18:00+08:00 SOP Need >5GB next-10 audit: remains Need >5GB. Runner audit is not_configured and current backend inventory found no FAMPNN source/runtime. Needs Python 3.10 PyTorch/Torch-Geometric/Hydra runtime and upstream Slurm benchmark. 2026-07-18T21:37:01+08:00 SOP Need >5GB batch source-staging: Official repo cloned to backend /media/nik/seagate_nik/bio_server/repos/fampnn at 58141f0 (296MB). Repo includes weights under weights/, but runtime needs Python 3.10, PyTorch GPU, Torch-Geometric, Hydra/OmegaConf, torchtyping, timm, natsort, einops, pandas, and tqdm. No SIF/benchmark yet. 2026-07-18T22:34:12+08:00 SOP Need >5GB continuation batch: No new runtime build completed in this continuation. Source and bundled weights remain staged; next production step remains Python3.10 PyTorch2.4.1 CUDA12.1 + PyG/Hydra SIF and sequence-design/packing benchmark.