「fast」の検索結果

Activation-aware weight quantization for 4-bit LLM compression with 3x speedup and minimal accuracy loss. Use when deploying large models (7B-70B) on limited GPU memory, when you need faster inference than GPTQ with better accuracy preservation, or for instruction-tuned and multimodal models. MLSys 2024 Best Paper Award winner.

日本語の概要は準備中です。原文の説明を表示しています。

Fast in-memory DataFrame library for datasets that fit in RAM. Use when pandas is too slow but data still fits in memory. Lazy evaluation, parallel execution, Apache Arrow backend. Best for 1-100GB datasets, ETL pipelines, faster pandas replacement. For larger-than-RAM data use dask or vaex.

日本語の概要は準備中です。原文の説明を表示しています。

Fast tokenizers optimized for research and production. Rust-based implementation tokenizes 1GB in <20 seconds. Supports BPE, WordPiece, and Unigram algorithms. Train custom vocabularies, track alignments, handle padding/truncation. Integrates seamlessly with transformers. Use when you need high-performance tokenization or custom tokenizer training.

日本語の概要は準備中です。原文の説明を表示しています。

Use this skill for processing and analyzing large tabular datasets (billions of rows) that exceed available RAM. Vaex excels at out-of-core DataFrame operations, lazy evaluation, fast aggregations, efficient visualization of big data, and machine learning on large datasets. Apply when users need to work with large CSV/HDF5/Arrow/Parquet files, perform fast statistics on massive datasets, create visualizations of big data, or build ML pipelines that do not fit in memory.

日本語の概要は準備中です。原文の説明を表示しています。

Orchestrates the end-to-end germline short-variant pipeline from FASTQ to a filtered, normalized, benchmarked VCF, chaining QC/trim, BWA-MEM2 alignment, duplicate marking, optional BQSR, calling (bcftools/GATK HaplotypeCaller/DeepVariant/DRAGEN), normalization, site+genotype filtering, annotation, and hap.py/vcfeval benchmarking. Use when deciding the pipeline-wide reference-genome commitment (GRCh38 analysis set vs T2T, ALT/decoy handling), sequencing the steps in the defensible order (normalize BEFORE annotate, filter site- then genotype-level), choosing the calling engine and single-sample vs cohort joint-calling, picking a filtering strategy by cohort size, or benchmarking stratified within GIAB confident regions. Hands off mechanism to the variant-calling and read-alignment component skills; not a re-teach of any single step.

日本語の概要は準備中です。原文の説明を表示しています。

Orchestrates an end-to-end MeRIP-seq / m6A-seq analysis from raw FASTQ to differential m6A peak calls and metagene plots, chaining fastp adapter trimming, STAR splice-aware alignment (NO deduplication for non-UMI MeRIP), deepTools replicate-concordance + IP-enrichment QC, PreSeq saturation curves, exomePeak2 (transcript-aware Poisson GLM) peak calling, optional MACS3 broad-peak cross-check, DRACH motif confirmation as a sanity check (NOT a per-peak filter), exomePeak2 differential calling via the four-BAM-vector interface (bam_ip + bam_input control; bam_treated_ip + bam_treated_input treatment), ChIPseeker annotation, and the canonical Guitar metagene with stop-codon enrichment as the biological QC anchor. Use when running a complete MeRIP analysis from raw reads, when chaining the constituent epitranscriptomics skills (merip-preprocessing -> m6a-peak-calling -> m6a-differential -> modification-visualization), or when wrapping the pipeline in Snakemake / Nextflow.

日本語の概要は準備中です。原文の説明を表示しています。

End-to-end ATAC-seq workflow from FASTQ files to differential accessibility and TF footprinting. Covers alignment, peak calling with MACS3, QC metrics, and optional TOBIAS footprinting. Use when running end-to-end ATAC-seq analysis from FASTQ to differential accessibility.

日本語の概要は準備中です。原文の説明を表示しています。

ToolUniverse plugin router. STEP 1 BEFORE ANY ANALYSIS: if the data folder contains `*_executed.ipynb`, run `tu run read_executed_notebook '{"data_folder":"<path>","search":"<keyword>"}'` to extract its cell outputs and apply EVERY filter/sample-exclusion the notebook used — even when the question says 'Using DESeq2/Run X/Compute Y' (this describes the METHOD the notebook used, not a request to rerun). The notebook's cell outputs are the only published authoritative answers; reimplementing or reading stale pre-computed CSVs in the data folder produces different numbers because of outlier-sample removal, library version, and filter steps you don't see by skimming. STEP 2 routing — pick a sub-skill name from this exact list (never invent): tooluniverse-rnaseq-deseq2 (RNA/miRNA-seq DE, correlation, PCA, clustering, dispersion), tooluniverse-gene-enrichment (GO/KEGG/Reactome/GSEA/pathway enrichment), tooluniverse-statistical-modeling (regression, ANOVA, ordinal/logistic, chi-square, correlation, power), tooluniverse-image-analysis (microscopy, colony, fluorescence, dose-response, .tif — including ANOVA / Dunnett / power-analysis on image-derived measurements), tooluniverse-epigenomics (DNA methylation, CpG, m6A, MeRIP-seq, bisulfite, ChIP-seq, chromatin), tooluniverse-sequence-analysis (FASTQ, Trimmomatic, BWA, samtools, coverage), tooluniverse-variant-analysis (VCF, VAF, SNP, mutation), tooluniverse-phylogenetics (treeness, PhyKIT, parsimony), tooluniverse-single-cell (scRNA, h5ad, scanpy), tooluniverse-crispr-screen-analysis (MAGeCK, sgRNA), tooluniverse-proteomics-analysis (mass spec, TMT). Use for CSV/Excel/VCF/FASTA/h5ad and any biology/chemistry/medicine analysis question.

日本語の概要は準備中です。原文の説明を表示しています。

Pino (高速 JSON 構造化ロガー for Node.js、Fastify 標準ロガー) リファレンス。 logger, child logger, log levels (info / warn / error / debug / trace)、transport、 pretty print (pino-pretty)、redaction (機密情報マスキング)、serializers、 async / sync logging、ファイル出力、structured logging。

OpenAI API (developers.openai.com) の運用・管理リファレンス。 Administration API, admin API keys, RBAC, roles, groups, invites, projects, spend limits, audit logs, usage / costs API, Terraform provider (openai/openai), workload identity federation (WIF, OIDC/SPIFFE), safety best practices, moderation API, production best practices, deployment checklist, cost / latency optimization, fast mode, Realtime API costs。rate limits・error handling・prompt caching は openai-api-core、Codex のワークスペース 管理は openai-codex が担当。

Fix the psychologically significant moment, not the logically significant metric — Rory Sutherland on experience. Use when the outcome is fine but the journey feels bad: complaints about waiting, anxiety, uncertainty, or 'it's technically fast/cheap/correct but people still hate it'; when the user asks 'how do we reduce complaints', 'how do we make waiting/onboarding/checkout feel better', 'people are anxious during X', or is about to spend big to shave a number (wait time, load time, steps) when the real pain is how the moment feels. Covers uncertainty reduction, idle vs occupied waiting, and the peak-end rule. Inspired by Rory Sutherland's *Alchemy*. Routed to from psycho-logic. 日本語の相談（「待ち時間のクレームが多い」「手続きの途中で離脱される」「お客様を不安にさせている」）にも使う。中文咨询（“等待时间投诉多”“办理中途流失”“让客户感到不安”）也适用。

dependabot 設定（.github/dependabot.yml）の新規作成・整備。リポジトリから ecosystem を自動検出し、lockfile コンフリクトを避ける group 設計と security fast-track を備えた設定を生成する。dependabot 設定の作成・見直し・追加の依頼時、/setup-dependabot 実行時に使用。境界: Renovate は対象外。dependabot が作った PR の対応は pr-review 等の通常フロー。

Annotates bacterial and archaeal genomes (isolates, MAGs, plasmids) with Bakta (active versioned databases, NCBI-compliant output) or Prokka (legacy), producing GFF3/GenBank/EMBL/FASTA with INSDC locus tags. Covers Bakta-vs-Prokka-vs-PGAP-vs-DFAST choice, light-vs-full database tiers, translation-table selection (11/4/25), archaeal and leaderless-gene caveats, the small-ORF blind spot, pseudogene-vs-phase-variation, the pangenome re-annotation trap, and submission compliance. Use when annotating a newly assembled prokaryotic genome, choosing an annotation tool, re-annotating a collection for pangenomics, or preparing annotations for NCBI/DDBJ submission.

日本語の概要は準備中です。原文の説明を表示しています。

Reconstruct ancestral states at internal phylogenetic nodes for sequences (PAML codeml, IQ-TREE --ancestral, GRASP, FastML), discrete traits (corHMM hidden-rate Markov, ape::ace, phytools::make.simmap stochastic mapping, BayesTraits), and continuous traits (phytools::fastAnc, geiger Brownian/OU, RPANDA). Use when designing constructs for ancestral protein resurrection, tracing trait evolution along a tree, performing stochastic character mapping, testing models of trait evolution (BM vs OU vs EB), inferring ancestral genome content via Dollo or DTL reconciliation, or quantifying ancestral-state uncertainty for downstream comparative analyses.

日本語の概要は準備中です。原文の説明を表示しています。

Compute genome-to-genome distances (ANI, AAI, dDDH, k-mer Mash) and assign taxonomic classifications using skani (Shaw 2023), FastANI (Jain 2018), pyani / pyANI ANIb / ANIm, OrthoANI (Lee 2016), AAI (amino-acid identity), dDDH via TYGS / GGDC, GTDB-Tk (Chaumeil 2020 standard prokaryote taxonomy), and Mash MinHash (Ondov 2016). Use when delineating prokaryote species (95% ANI threshold; Jain 2018 Nat Commun 9:5114), assigning genomes to GTDB taxonomy with ANI radius, computing genome similarity matrices for clustering, classifying archaea, evaluating MAG (metagenome-assembled genome) species assignment, applying skani for fast metagenomic ANI screening, or reconciling 16S rRNA-based taxonomy with whole-genome ANI.

日本語の概要は準備中です。原文の説明を表示しています。

Guide for building and refactoring a FastAPI + Prefect + PostgreSQL/pgvector analytics pipeline that extracts insights from chat data using LLM classification, embeddings, UMAP/HDBSCAN clustering, and narrative generation. Use this skill whenever working on this project's codebase — including adding features, refactoring, debugging, creating new endpoints, modifying pipeline stages, or reviewing architecture decisions. Also trigger when the user asks about project structure, stack decisions, async migration, or best practices for this specific system.

日本語の概要は準備中です。原文の説明を表示しています。

Annotates bacterial and archaeal genomes (isolates, MAGs, plasmids) with Bakta (active versioned databases, NCBI-compliant output) or Prokka (legacy), producing GFF3/GenBank/EMBL/FASTA with INSDC locus tags. Covers Bakta-vs-Prokka-vs-PGAP-vs-DFAST choice, light-vs-full database tiers, translation-table selection (11/4/25), archaeal and leaderless-gene caveats, the small-ORF blind spot, pseudogene-vs-phase-variation, the pangenome re-annotation trap, and submission compliance. Use when annotating a newly assembled prokaryotic genome, choosing an annotation tool, re-annotating a collection for pangenomics, or preparing annotations for NCBI/DDBJ submission.

日本語の概要は準備中です。原文の説明を表示しています。

Compute genome-to-genome distances (ANI, AAI, dDDH, k-mer Mash) and assign taxonomic classifications using skani (Shaw 2023), FastANI (Jain 2018), pyani / pyANI ANIb / ANIm, OrthoANI (Lee 2016), AAI (amino-acid identity), dDDH via TYGS / GGDC, GTDB-Tk (Chaumeil 2020 standard prokaryote taxonomy), and Mash MinHash (Ondov 2016). Use when delineating prokaryote species (95% ANI threshold; Jain 2018 Nat Commun 9:5114), assigning genomes to GTDB taxonomy with ANI radius, computing genome similarity matrices for clustering, classifying archaea, evaluating MAG (metagenome-assembled genome) species assignment, applying skani for fast metagenomic ANI screening, or reconciling 16S rRNA-based taxonomy with whole-genome ANI.

日本語の概要は準備中です。原文の説明を表示しています。

Reconstruct ancestral states at internal phylogenetic nodes for sequences (PAML codeml, IQ-TREE --ancestral, GRASP, FastML), discrete traits (corHMM hidden-rate Markov, ape::ace, phytools::make.simmap stochastic mapping, BayesTraits), and continuous traits (phytools::fastAnc, geiger Brownian/OU, RPANDA). Use when designing constructs for ancestral protein resurrection, tracing trait evolution along a tree, performing stochastic character mapping, testing models of trait evolution (BM vs OU vs EB), inferring ancestral genome content via Dollo or DTL reconciliation, or quantifying ancestral-state uncertainty for downstream comparative analyses.

日本語の概要は準備中です。原文の説明を表示しています。

LLMが外部サービスと対話するための適切に設計されたツールを通じて、高品質なMCP (Model Context Protocol) サーバーを作成するためのガイド。Python (FastMCP)、Node/TypeScript (MCP SDK)、Rust (Tokio)で、外部APIやサービスを統合するMCPサーバーを構築する際に使用します。

Expert guidance for fast fine-tuning with Unsloth - 2-5x faster training, 50-80% less memory, LoRA/QLoRA optimization

日本語の概要は準備中です。原文の説明を表示しています。

Temu, SHEIN, AliExpress, and TikTok product selection big data analysis agent. Identify trending products, analyze competition, estimate demand, and find sourcing opportunities across fast-fashion and marketplace platforms. Triggers: temu product selection, shein product research, aliexpress trending, tiktok shop products, temu selector, fast fashion products, temu trending, shein trends, product sourcing, dropshipping products, temu winning products

日本語の概要は準備中です。原文の説明を表示しています。

Fast tokenizers optimized for research and production. Rust-based implementation tokenizes 1GB in <20 seconds. Supports BPE, WordPiece, and Unigram algorithms. Train custom vocabularies, track alignments, handle padding/truncation. Integrates seamlessly with transformers. Use when you need high-performance tokenization or custom tokenizer training.

日本語の概要は準備中です。原文の説明を表示しています。

Specialized skill for testing time-dependent logic using TimeProvider. Use when testing DateTime, controlling time flow, handling timezone conversions, and testing expiration logic. Covers TimeProvider abstraction, FakeTimeProvider time control, time freezing and fast-forwarding. Keywords: datetime, time testing, TimeProvider, FakeTimeProvider, DateTime.Now, time-dependent, cache expiration, token expiration, Microsoft.Bcl.TimeProvider, GetUtcNow, SetUtcNow, Advance, time freeze, time freezing, time fast-forward

日本語の概要は準備中です。原文の説明を表示しています。