Claude Skills by VectorSpaceLab

"Construct, review, and troubleshoot D-FINE train.py commands for

"Use when working with Dagster OSS: assets, jobs, Definitions,

"Model, validate, test, and troubleshoot Dagster assets, jobs,

"Define, debug, and test Dagster schedules, sensors, asset sensors,

"Use when a coding agent needs to run local Dagster OSS CLI

"Create and troubleshoot Dagster component-ready projects,

"Use this sub-skill when implementing or debugging Dagster Pythonic

"Use when a coding agent needs to configure or troubleshoot Dagster

"Use when a coding agent needs to start or debug dagster-webserver,

"Integrate Python and non-Python external processes with Dagster

"Use when a coding agent edits the Dagster OSS repository itself:

"Operate lucidrains DALLE-pytorch for DALL-E style VAE training,

"Train and resume DALLE-pytorch transformer workflows with

"Choose and troubleshoot DALLE-pytorch CUDA, DeepSpeed, Horovod,

"Generate images or text with DALLE-pytorch checkpoints, prime with

"Train and inspect DALLE-pytorch discrete VAE workflows, VAE

"Use the DALLE2-pytorch package for DALL-E 2 style text-to-image

"Use DALLE2-pytorch dataloaders, WebDataset and embedding layouts,

"Use DALLE2-pytorch public generation and model APIs: DALLE2,

"Configure and launch DALLE2-pytorch decoder and diffusion-prior

Training, inference, and deployment workflows for DAMO-YOLO object detection.

"Export DAMO-YOLO models to ONNX or TensorRT, plan partial INT8

"Run DAMO-YOLO image, video, and camera inference with Torch, ONNX,

"Training, fine-tuning, evaluation, distillation, custom COCO

"Routes Darkflow object-detection, training, and export workflows

"Guides Darkflow image, video, JSON, Python API, and protobuf

"Guides Darkflow custom dataset preparation, YOLO config edits,

"Use when building, debugging, testing, or maintaining Plotly Dash

"Use for Plotly Dash app layouts, callbacks, pages, assets,

"Use for Dash component wrapper generation, built-in component

"Use for Dash backend selection, async callbacks, WebSocket

"Use for Dash test selection, browser fixtures, build/lint

"Use this repo skill for Dask, the Python parallel computing

"Use when working with Dask Array: chunked NumPy-like arrays,

"Use for Dask Bag, dask.bytes, text/byte IO, JSON-like records,

"Use this Dask sub-skill for configuration files and environment

"Use this sub-skill for Dask delayed, task graph inspection,

"Use this Dask sub-skill for Dask DataFrame creation,

"Use when you need to design DataDesigner configs, run

"Helps agents use the Data Designer CLI, config subcommands,

"Design, inspect, validate, and serialize DataDesigner configs:

"Run DataDesigner through the Python API for validate,

"Work with DataDesigner plugin descriptors, catalog package

"Use when adapting DataDesigner tutorial notebooks, documented

"Data-Juicer repo router for local recipes, Ray recovery, and

"Data-Juicer Ray execution, partitioning, checkpointing, recovery,

Local Data-Juicer recipe processing, analysis, export, and utility workflows.

"Data-Juicer FastAPI service, MCP server, operator discovery, and

"Use the DataSciencePython tutorial/example collection through

"Safe, modern Kaggle-style SVM and logistic-regression workflows