
Claude Skills by VectorSpaceLab
github.com/VectorSpaceLab"This skill guides Faiss CPU/GPU acceleration, backend-gated
"This skill teaches a Researcher to compose Faiss indexes, preserve
"Guides dense-vector metric selection, index construction,
"The persistence-and-evaluation sub-skill routes Faiss index
"This skill teaches a Researcher to train, compose, inspect, and
"Operate FastReID setup, configuration, datasets, model/inference
"Operate FastReID dataset registries, built-in layouts, custom
"Operate FastReID deployment export surfaces, optional
"Operate FastReID model registries, safe CPU construction, feature
"Set up FastReID from a source-only checkout, verify imports, and
"Operate FastReID training and evaluation launch commands,
"Guides TensorFlow Fast Style Transfer training, image stylization,
"Guides Fast Style Transfer evaluate.py still-image and
"Plans and validates Fast Style Transfer style.py training runs,
"Guides Fast Style Transfer transform_video.py and ffwd_video
"Route fastdup workflows for visual dataset curation,
"Guide fastdup workflows for local image cleanup, duplicate
"Guide fastdup workflows for video inputs, tar or zip archives,
"Guide fastdup workflows for embeddings, feature-vector search,
"Guide fastdup workflows for labeled image datasets, object
"Use the faster-whisper package for CTranslate2-backed Whisper
"Use faster-whisper transcription APIs for ASR, batched inference,
"Use fastMRI for MRI reconstruction data loading, MRI operators,
"Load and inspect fastMRI HDF5 data, datasets, masks, transforms,
"Prepare fastMRI reconstructions for validation, metric evaluation,
"Build fastMRI PyTorch Lightning training and test workflows with
"Instantiate, inspect, and debug fastMRI reconstruction model
"Use fastMRI complex tensor math, centered FFT/IFFT, RSS coil
Guides FastVideo video, image, and audio generation, config-first inference, serving, training, distillation, evaluation, and performance workflows.
Guides FastVideo DMD and self-forcing distillation, Attn-QAT, LoRA extraction/merge/verification, and checkpoint conversion planning.
Guides FastVideo media metrics, benchmark configuration, quality regression checks, and honest performance comparisons.
Guides FastVideo typed Python and config-first CLI generation, model inputs, outputs, offload, attention backends, quantization, and compilation.
Guides FastVideo HTTP and WebSocket serving, typed request contracts, health checks, session state, and server configuration.
Handles FastVideo installation, platform and backend selection, model registry lookup, presets, configuration schemas, and environment diagnosis.
Guides FastVideo dataset layouts, latent preprocessing, modular YAML training, legacy recipe boundaries, and resource-safe training setup.
"Operate FATE 2.2.0 federated-learning deployments, FateFlow
"Inspect FATE component CLI commands, component descriptors, task
"Install and deploy FATE through the documented PyPI, standalone
"Run and author service-free local FATE launchers with
"Service-backed FATE-Client Pipeline workflows for data upload,
"Routes FCOS object-detection repo tasks for inference demos,
"Guides FCOS YAML config selection, MODEL.FCOS options, dataset
"Guides FCOS image inference, installed CLI use, public FCOS API
"Guides FCOS source internals, model/loss/postprocess components,
"Guides FCOS ONNX export command construction, output tensor
"Builds and troubleshoots FCOS training, evaluation, distributed
"Use for Feast feature store tasks: feature repositories,
"Model Feast entities, fields, data sources, feature views,
"Create and operate Feast feature repositories with
"Use when selecting or configuring Feast optional stores,