Claude Skills by VectorSpaceLab

"Use evaluate.evaluator and EvaluationSuite to evaluate model

"Create, validate, and publish custom Hugging Face Evaluate

"Compute evaluate metrics, comparisons, and measurements with

"Discover, inspect, and load Hugging Face Evaluate metrics,

"Use this skill for EverOS local-first Markdown memory service

"Use this sub-skill for EverOS Markdown storage layout, cascade

"Use this sub-skill for EverOS knowledge document upload, topic

"Use this sub-skill for EverOS HTTP memory add, flush, search, get,

"Use this sub-skill for EverOS logging, Prometheus metrics,

"Use this sub-skill to install EverOS, generate and inspect

"Routes evo trajectory-evaluation workflows for APE/RPE, trajectory

Routes evo package-info, settings, logfile, and IPython shell workflows.

"Routes evo_ape and evo_rpe workflows, metric helper APIs,

"Routes evo programmatic usage, notebook workflows, plotting, and

"Routes evo_res workflows, saved-result comparison, table export,

"Routes evo_traj workflows, trajectory file formats, converters,

"Use ExecuTorch to export PyTorch models to edge runtime artifacts,

"Choose and configure ExecuTorch backends/delegates, including CPU,

"Measure, compare, and reduce ExecuTorch binary size using

"Develop, export, build, and test ExecuTorch Cortex-M/CMSIS-NN

"Export PyTorch models to ExecuTorch .pte/.ptd programs and

"Plan ExecuTorch LLM export and on-device runner workflows,

"Profile and debug ExecuTorch execution with ETRecord, ETDump,

"Build, export, test, and debug ExecuTorch Qualcomm QNN backend

"Install and build ExecuTorch from source, including Python package

"Routes face-alignment installation, import checks,

"Routes face detector backend selection and backend-specific

"Routes face landmark prediction on single images, batches, and

"Use face.evoLVe for high-performance face recognition workflows

"Prepare and validate face.evoLVe identity-folder datasets,

"Align face identity folders with MTCNN detection, landmark

"Extract face.evoLVe embeddings from trained PyTorch checkpoints

"Use face.evoLVe PaddlePaddle training, quantization, Paddle

"Configure and inspect face.evoLVe PyTorch training, validation,

"Use face_recognition to detect faces, extract landmarks and

"Use davidsandberg/facenet for TensorFlow 1.x face recognition

"Prepare Facenet class-folder datasets, validate LFW pair inputs,

"Use Facenet embeddings for face comparison, SVM classifiers,

"Evaluate Facenet embeddings with LFW pair files, ROC/VAL/FAR

"Load Facenet checkpoint directories or frozen graphs, inspect

"Train Facenet models with softmax or triplet loss, model

"Use Fairlearn to assess group fairness, compute disparity metrics,

"Use Fairlearn adversarial fairness estimators with PyTorch or

"Use Fairlearn assessment APIs for MetricFrame, group fairness

"Use Fairlearn built-in dataset loaders, schema notes, cache

"Install, inspect, and troubleshoot Fairlearn runtimes, optional

"Use Fairlearn ThresholdOptimizer and threshold-optimizer plots to

"Use Fairlearn preprocessing mitigation with CorrelationRemover and

"Use Fairlearn reductions mitigation with ExponentiatedGradient,

"Guides Researchers through Faiss installation, dense and binary