Claude Skills by VectorSpaceLab

Prepare XLNet pretraining corpora, TFRecords, and GPU/TPU command plans.

"Use XLNet TensorFlow graph APIs, config objects, tokenization

"Operate XLNet RACE multiple-choice reading-comprehension training

"Operate XLNet SQuAD 1.1/2.0 preprocessing, fine-tuning,

"Guide Researcher agents through XrayGLM Chinese medical

"Validate and transform XrayGLM image, prompt, and label data with

"Prepare and validate XrayGLM supervised multimodal fine-tuning

"Operate checkpoint-backed XrayGLM inference from the CLI or Gradio

"Use XTuner for large-model SFT, pretraining, MLLM fine-tuning,

"Operate XTuner legacy command routing, old config-zoo discovery,

"Validate and prepare XTuner SFT, MLLM, pretraining, and RL JSONL

"Select and troubleshoot XTuner V1 model configs, MoE/VLM backends,

"Plan and troubleshoot XTuner RL and GRPO workflows with Ray

"Plan and troubleshoot XTuner V1 SFT, pretraining, and multimodal

"Use Yellowbrick visual diagnostics for scikit-learn models,

"Use Yellowbrick classification diagnostics for scikit-learn

"Use Yellowbrick clustering diagnostics and model-selection

"Use Yellowbrick experimental contrib visualizers, missing-data

"Use Yellowbrick feature and target visualizers for feature

"Use Yellowbrick regression diagnostics for residuals, prediction

"Use Yellowbrick dataset loaders, cache controls, and text

"Operate YiVal prompt/model evaluation workflows, including YAML

"Implement and register custom YiVal readers, wrappers, evaluators,

"Configure YiVal evaluators, human ratings, AHP selection, and

"Configure YiVal data generators and prompt variation generators

"Run YiVal experiments through the CLI or Python APIs, interpret

"Install YiVal, inspect CLI commands, create/validate experiment

"Guides YOLOP multi-task driving perception workflows for BDD100K

"Prepares and validates BDD100K-style YOLOP data roots, detection

"Guides YOLOP ONNX export, ONNXRuntime inference validation,

"Guides YOLOP PyTorch demo inference, checkpoint loading,

"Guides YOLOP training, validation, evaluation metrics, staged

"Operate Ultralytics YOLOv3 detection workflows: inference,

"Export YOLOv3 weights and choose deployment formats across

"Run YOLOv3 inference through detect.py, PyTorch Hub, and

"Inspect and modify YOLOv3 model YAMLs, anchors, Detect heads,

"Maintain the YOLOv3 repository under its CI, style, compatibility,

"Plan and debug YOLOv3 training, custom dataset YAMLs, checkpoints,

"Run and interpret YOLOv3 validation, mAP metrics, val.py tasks,

"Use this skill for clone-run Ultralytics YOLOv5 workflows:

"Use this YOLOv5 sub-skill for image-classification training,

"Use this YOLOv5 sub-skill for object-detection training,

"Use this YOLOv5 sub-skill for model export, deployment formats,

"Use this YOLOv5 sub-skill for instance-segmentation training,

"Use this YOLOv5 sub-skill for the Flask REST API example, request

"Operate YOLOv7-d2, a Detectron2-based repository for YOLO-family,

"Export YOLOv7-d2 models to ONNX or TorchScript, inspect

"Run and troubleshoot YOLOv7-d2 PyTorch demo inference,

"Choose, validate, and launch YOLOv7-d2 Detectron2 configs, custom

"Use YOLOX for object-detection inference, training/data