Routes 3D ResNets PyTorch video action-recognition workflows across training, inference, and data preparation.
The English summary is being prepared. Showing the original description.
Use the Folding@home coronavirus repository conventions to curate molecular structures, prepare bounded OpenMM systems, and preserve target, publication, and provenance notes.
The English summary is being prepared. Showing the original description.
See how to installReview the instructions the agent will receive before you install.
Use this repo-specific operating skill for reusable work derived from the Folding@home coronavirus preparation repository. It covers structure curation, OpenMM preparation/equilibration mechanics, and project-context notes. It does not run Folding@home clients, network acquisition, long production trajectories, or medical/therapeutic interpretation.
For a new simulation input, establish source-structure identity and note provenance first; curate and validate the coordinate topology; choose and document the force-field/ligand parameter path; prepare a small CPU OpenMM system; run only a bounded minimization/equilibration smoke check; inspect energy, atom count, box, and output bundle; then record exact parameters, versions, warnings, and limitations. A later Researcher may expand the run only with an explicit scientific and compute budget.
Do not treat a filename, a successful PDB parse, a short CPU step, or XML serialization as proof of biological identity, force-field suitability, stable long-timestep dynamics, binding, efficacy, or reproducibility across platforms.
The source project documents rapidly prepared inputs and commonly uses equilibrated_4fs.pdb as the Folding@home handoff. Reusable generated helpers use safer explicit arguments and short defaults instead of historical hard-coded paths or multi-nanosecond defaults. A 4 amu hydrogen mass and 4/5 fs continuation are historical workflow conventions that must be disclosed and validated for the actual system, not universal recommendations.
The required backend for this selected scope is CPU OpenMM. CUDA is optional: platform enumeration succeeded in inspection, but context creation was limited by CUDA_ERROR_UNSUPPORTED_PTX_VERSION. Do not claim CUDA verification or require it for a CPU result.
Use repo-provenance.md to anchor claims to the source revision and evidence paths. This generated skill is intentionally left in the repository-local output tree and is not imported into the managed DisCo skill library.
No reviews yet. Tell others what you thought.
No login needed. Share how it worked for you, ask a question, or note where it did not run.
No comments yet. Be the first to write one.
Routes 3D ResNets PyTorch video action-recognition workflows across training, inference, and data preparation.
The English summary is being prepared. Showing the original description.
Guide 3DDFA Python inference, geometry rendering, training/evaluation, and optional C++ ONNX workflows for 3D dense face alignment.
The English summary is being prepared. Showing the original description.
Routes 3DDFA_V2 face-alignment setup, still-image demos, video tracking, and ONNX benchmarking workflows.
The English summary is being prepared. Showing the original description.
Operate AB3DMOT 3D multi-object tracking workflows for KITTI and nuScenes data, tracking, evaluation, and visualization.
The English summary is being prepared. Showing the original description.
Use Hugging Face Accelerate for PyTorch training-loop migration, distributed launch/configuration, DeepSpeed/FSDP/TPU backend setup, big-model inference/offload, checkpointing, tracking, and troubleshooting.
The English summary is being prepared. Showing the original description.
Route Acme reinforcement-learning framework tasks across core loops, replay/data, JAX agents, and TensorFlow/Sonnet agents.
The English summary is being prepared. Showing the original description.