Claude Skills by VectorSpaceLab

"Use Time-Series-Library (TSLib) to configure, run, debug, and

"Prepare Time-Series-Library data layouts, choose safe run.py

"Configure, run, and debug Time-Series-Library long-term,

"Navigate Time-Series-Library model files, optional Mamba and large

"Configure and debug Time-Series-Library imputation, reconstruction

"Use TimeMixer to inspect the model API, build forecasting or

"Validate and explain TimeMixer dataset layouts, file names,

Construct and adapt TimeMixer forecasting experiment commands.

"Use and debug the TimeMixer Model API, PDM/FMM architecture,

"Use and debug TimeMixer imputation, anomaly detection, and

"Use PyTorch Image Models (timm) for model discovery, pretrained

"Use this sub-skill when a user needs to run bounded timm

"Generate safe, correct train.py, validate.py, inference.py, and

"Build timm data configs, datasets, transforms, loaders, ImageNet

"Build ONNX export and validation commands, package/load timm

"Use timm reusable layers, classifier heads, pooling,

"Discover timm model names, instantiate models, load pretrained or

"Compose timm API-level training loops with optimizer and scheduler

"Use Torch Points3D for point-cloud deep learning APIs,

"Use Torch Points3D dataset factories, point-cloud data layouts,

"Use Torch Points3D high-level model factories, application APIs,

"Use Torch Points3D registration datasets, configs, pair/fragment

"Use Torch Points3D Hydra training, evaluation, Trainer,

"Use this skill for Torch-TensorRT tasks: compiling PyTorch models

"Use this sub-skill for Torch-TensorRT source builds, package

"Use this sub-skill for Torch-TensorRT model compilation, dynamic

"Use this sub-skill for Torch-TensorRT deployment artifacts, Triton

"Use this sub-skill for Torch-TensorRT dryrun analysis, Debugger

"Use this sub-skill for Torch-TensorRT runtime performance

"Route TorchDrug graph learning, drug discovery, molecular ML,

"Construct, batch, split, mask, validate, and troubleshoot

"Use TorchDrug for knowledge graph reasoning, triple datasets,

"Choose TorchDrug graph neural network layers/models, readouts,

"Plan TorchDrug molecular property prediction, molecular

"Plan TorchDrug protein sequence, structure, contact prediction,

"Assemble TorchDrug models and tasks into core.Engine training,

"Use for TorchGeo geospatial deep learning work: datasets,

"Use for TorchGeo repository edits, style/type conventions, adding

"Use for TorchGeo Lightning DataModules, CLI/config runs, task

"Use for TorchGeo GeoDataset/NonGeoDataset usage, dataset

"Use for TorchGeo model registry, pretrained weights, model

"Use TorchIO 2.0 for medical image loading, preprocessing,

"Use TorchIO's command-line interface and I/O integration surfaces

"Construct and validate TorchIO images, subjects/studies, affines,

"Design TorchIO patch-based training queues and dense patch

"Build TorchIO 2.0 preprocessing and augmentation pipelines with

"Use TorchMetrics to choose, inspect, and combine metric families

"Use TorchMetrics audio/speech quality metrics and no-download text

"Use TorchMetrics classification, regression, retrieval,

"Compose, transform, track, bootstrap, and plot TorchMetrics