1185 个 · 按相关度排序
Use Transformers.js to run state-of-the-art machine learning models directly in JavaScript/TypeScript. Supports NLP (text classification, translation, summarization), computer vision (image classification, object detection), audio (speech recognition, audio classification), and multimodal tasks. Works in browsers and server-side runtimes (Node.js, Bun, Deno) with WebGPU/WASM using pre-trained models from Hugging Face Hub.
中文简介正在准备中。当前显示原文说明。
Expert en transformation digitale (strategy, change management, technology adoption, maturity assessment)
中文简介正在准备中。当前显示原文说明。
Corrects fluorophore spillover (conventional compensation) or spectral overlap (spectral unmixing) and applies variance-stabilizing transforms (logicle/biexponential, arcsinh, log) for flow and mass cytometry. Covers spillover-matrix estimation from single-stain controls, AutoSpill, the spillover spreading matrix and why panel design (not compensation) bounds resolution, compensate-then-transform ordering, and arcsinh cofactor choice (5 for CyTOF, ~150 for fluorescence, per-channel via flowVS). Use when correcting spectral overlap, preparing data for gating/clustering, choosing logicle vs arcsinh, deciding a cofactor, or distinguishing compensation from spectral unmixing.
中文简介正在准备中。当前显示原文说明。
Plan and execute database migrations, data transformations, and system migrations safely with rollback strategies and data integrity validation. Use when migrating databases, transforming data schemas, moving between database systems, implementing versioned migrations, handling data transformations, ensuring data integrity, or planning zero-downtime migrations.
中文简介正在准备中。当前显示原文说明。
Corrects fluorophore spillover (conventional compensation) or spectral overlap (spectral unmixing) and applies variance-stabilizing transforms (logicle/biexponential, arcsinh, log) for flow and mass cytometry. Covers spillover-matrix estimation from single-stain controls, AutoSpill, the spillover spreading matrix and why panel design (not compensation) bounds resolution, compensate-then-transform ordering, and arcsinh cofactor choice (5 for CyTOF, ~150 for fluorescence, per-channel via flowVS). Use when correcting spectral overlap, preparing data for gating/clustering, choosing logicle vs arcsinh, deciding a cofactor, or distinguishing compensation from spectral unmixing.
中文简介正在准备中。当前显示原文说明。
dbt Core/Cloud data transformations, testing, documentation, and CI/CD. Activate on: dbt, data transformation, analytics engineering, ref, source, staging model, mart, dbt test. NOT for: orchestration/scheduling (use airflow-dag-orchestrator), data warehouse tuning (use data-warehouse-optimizer).
中文简介正在准备中。当前显示原文说明。
Work with state-of-the-art machine learning models for NLP, computer vision, audio, and multimodal tasks using HuggingFace Transformers. This skill should be used when fine-tuning pre-trained models, performing inference with pipelines, generating text, training sequence models, or working with BERT, GPT, T5, ViT, and other transformer architectures. Covers model loading, tokenization, training with Trainer API, text generation strategies, and task-specific patterns for classification, NER, QA, summarization, translation, and image tasks. (plugin:scientific-packages@claude-scientific-skills)
中文简介正在准备中。当前显示原文说明。
Optimizes transformer attention with Flash Attention for 2-4x speedup and 10-20x memory reduction. Use when training/running transformers with long sequences (>512 tokens), encountering GPU memory issues with attention, or need faster inference. Supports PyTorch native SDPA, flash-attn library, H100 FP8, and sliding window attention.
中文简介正在准备中。当前显示原文说明。
通过改变提示词和输入表达，检验经API访问的大语言模型对安全过滤绕过尝试的响应，并比较多个模型的回答及保存Hermes设置。
使用稀疏自编码器将语言模型内部的激活分解为少量特征，研究模型学到的概念。涵盖预训练SAE分析、自定义训练、重建质量评估，以及利用特征调整输出的实验。
对本地或持续集成中发现的 Python 类型检查错误进行分类，通过类型收窄和注解修正等方式解决并重新验证，也处理新增类型检查范围的工作。
Connect agent tasks into a sequence where each step receives the previous output. Use custom prompts or predefined workflows to analyze code, plan refactoring, generate tests, optimize performance, or transform data through multiple stages.
中文简介正在准备中。当前显示英语简介。
Explain a trading model's prediction by ranking the input features that contributed to its signal. Returns a Markdown table and saves an attribution record with calculation settings, adding a digital signature when a signing key is available.
中文简介正在准备中。当前显示英语简介。
Assess tasks before execution to identify transformations eligible to skip LLM calls. Report the eligible share, recommended models for remaining tasks, and estimated upper costs, then save the routing results for later cost reports.
中文简介正在准备中。当前显示英语简介。
Connect agent tasks in sequence, passing each result into the next step. Build custom workflows or use predefined pipelines for code analysis, refactoring, testing, and optimization, with execution progress, timing, and consolidated results.
中文简介正在准备中。当前显示英语简介。
Manage Cloudinary images, videos, and files with natural language requests. Organize folders, configure upload rules, inspect assets, generate transformed versions, and check storage and bandwidth usage through Rube MCP and Composio.
中文简介正在准备中。当前显示英语简介。
Test your understanding of a phase in AI Engineering from Scratch through eight multiple-choice questions. Receive a score, explanations for incorrect answers, and specific lessons to review before continuing your studies.
中文简介正在准备中。当前显示英语简介。
Improve GSAP animation performance by reviewing animated properties, browser layout work, and simultaneous updates. Use transforms, grouped updates, reusable animations, and targeted cleanup to reduce stuttering in lists, mouse followers, and scroll effects.
中文简介正在准备中。当前显示英语简介。
Animate characters built from SVG parts using GSAP and CSS transforms. Coordinate gestures and facial expressions, keep joints connected, and control playback by frame in Remotion or validate browser previews with HyperFrames.
中文简介正在准备中。当前显示英语简介。
Clean messy tabular data through OpenRefine using command-line operations. Import files, inspect rows, apply reusable JSON transformation histories, and export cleaned CSV or TSV data, with session support for undoing and redoing changes.
中文简介正在准备中。当前显示英语简介。
Inspect and transform local DICOM medical imaging files, including metadata, image frames, and compression formats. Generate technical inventories and preview images, check installed codecs, and review pseudonymized derivatives without treating those checks as proof of de-identification.
中文简介正在准备中。当前显示英语简介。
Inspect supported local scientific files before modeling, summarize structure and missing values, and flag possible data leakage or outlier effects. Produces bounded aggregate reports with identifiers obscured and provides a structured analysis report template.
中文简介正在准备中。当前显示英语简介。
Refactor JavaScript using ES6+ features to make code easier to read and maintain. Covers asynchronous operations with async/await, modules, and functional patterns that transform data without changing the original values.
中文简介正在准备中。当前显示英语简介。
Screen AI application inputs and historical logs for prompt injection, where malicious instructions try to redirect the model. Combine pattern matching, text structure scoring, and a DeBERTa classifier to flag suspicious content and produce JSON results.
中文简介正在准备中。当前显示英语简介。