NetSolP

secondsPaperCodeProduction Ready

Solubility and usability based on protein language models

Readiness

Readiness: Production Ready
Smoke: passed
Artifacts: passed
Production: production_pass
Storage: >5
Strategy: validated; official DTU standalone bundle with CPU-only PyTorch/ONNX Runtime venv
Next: Expose model type, prediction type, and FASTA upload controls in UI; keep full 8.2GB model bundle on Seagate storage.

Official DTU netsolp-1.0.ALL.tar.gz downloaded (6,050,225,681 bytes); extracted trained ONNX models occupy about 8.2GB. CPU-only venv installed. Official sample FASTA and Slurm benchmark passed; Slurm job 231 completed in 3.160s with 9 artifacts.