1천건 · 관련도순
Compress large language models using knowledge distillation from teacher to student models. Use when deploying smaller models with retained performance, transferring GPT-4 capabilities to open-source models, or reducing inference costs. Covers temperature scaling, soft targets, reverse KLD, logit distillation, and MiniLLM training strategies.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Train Physics-Informed Neural Networks (PINNs) using DeepXDE. Solve forward and inverse PDE problems by embedding physics equations into the neural network loss function. Supports 1D/2D/3D, time-dependent, and parametric PDEs.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Provides guidance for performing causal interventions on PyTorch models using pyvene's declarative intervention framework. Use when conducting causal tracing, activation patching, interchange intervention training, or testing causal hypotheses about model behavior.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Provides guidance for training and analyzing Sparse Autoencoders (SAEs) using SAELens to decompose neural network activations into interpretable features. Use when discovering interpretable features, analyzing superposition, or studying monosemantic representations in language models.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Production-ready reinforcement learning algorithms (PPO, SAC, DQN, TD3, DDPG, A2C) with scikit-learn-like API. Use for standard RL experiments, quick prototyping, and well-documented algorithm implementations. Best for single-agent RL with Gymnasium environments. For high-performance parallel training, multi-agent systems, or custom vectorized environments, use pufferlib instead.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Reduce LLM size and accelerate inference using pruning techniques like Wanda and SparseGPT. Use when compressing models without retraining, achieving 50% sparsity with minimal accuracy loss, or enabling faster inference on hardware accelerators. Covers unstructured pruning, structured pruning, N:M sparsity, magnitude pruning, and one-shot methods.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
This skill should be used when working with genomic interval data (BED files) for machine learning tasks. Use for training region embeddings (Region2Vec, BEDspace), single-cell ATAC-seq analysis (scEmbed), building consensus peaks (universes), or any ML-based analysis of genomic regions. Applies to BED file collections, scATAC-seq data, chromatin accessibility datasets, and region-based genomic feature learning.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Evaluates LLMs across 60+ academic benchmarks (MMLU, HumanEval, GSM8K, TruthfulQA, HellaSwag). Use when benchmarking model quality, comparing models, reporting academic results, or tracking training progress. Industry standard used by EleutherAI, HuggingFace, and major labs. Supports HuggingFace, vLLM, APIs.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
High-performance reinforcement learning framework optimized for speed and scale. Use when you need fast parallel training, vectorized environments, multi-agent systems, or integration with game environments (Atari, Procgen, NetHack). Achieves 2-10x speedups over standard implementations. For quick prototyping or standard algorithm implementations with extensive documentation, use stable-baselines3 instead.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Quantizes LLMs to 8-bit or 4-bit for 50-75% memory reduction with minimal accuracy loss. Use when GPU memory is limited, need to fit larger models, or want faster inference. Supports INT8, NF4, FP4 formats, QLoRA training, and 8-bit optimizers. Works with HuggingFace Transformers.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Patterns for loading PDE simulation datasets (PDEBench, PhiFlow, JAX-CFD) from HDF5 files. Handles layout detection (single tensor vs separate variables), spatial/temporal downsampling, multi-variable systems, HuggingFace and DaRUS data sources, and efficient PyTorch DataLoader creation. Use when preparing PDE data for neural operator training.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
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.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Authors reproducible bioinformatics pipelines with Snakemake - rules wired by output-file pattern, wildcards and expand() for sample fan-out, checkpoints for runtime-unknown outputs, resource/retry escalation, and conda/container software deployment on HPC and cloud. Use when deciding rule-based (Snakemake) vs channel/dataflow (Nextflow) authoring; wiring rules by OUTPUT-file pattern rather than imperative order; using wildcards + expand() for sample fan-out and constraining them to stop silent mis-routing; adding checkpoints when the set of outputs is unknown until a step runs (dynamic DAG); diagnosing why a job reran (or did not) under the mtime-plus-provenance trigger set; escalating memory on retry for OOM-killed jobs; and porting a Snakemake 7 `--cluster`/remote-provider command to the Snakemake 8+ executor-plugin and storage-plugin model (snakemake-executor-plugin-slurm) with `--software-deployment-method`.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Standardizes molecular structures using ChEMBL chembl_structure_pipeline and RDKit rdMolStandardize covering sanitization, salt/solvent stripping, neutralization, tautomer canonicalization, stereochemistry standardization, mixture handling, and isotope normalization. Explicitly compares ChEMBL pipeline, canSARchem, and PubChem standardization choices. Use when preparing libraries for QSAR training, joining datasets across sources, deduplicating compound collections, or building canonical compound registries.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Plan AI fine-tuning projects with data preparation, training, and evaluation. TRIGGERS - Use when user needs help with ai-fine-tuning-plan related tasks.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Build DPO training pipelines for AI alignment. TRIGGERS - Use when user needs help with ai-dpo-trainer related tasks.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Build data pipelines for AI model training. TRIGGERS - Use when user needs help with ai-data-pipeline related tasks.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Curate and prepare datasets for AI training. TRIGGERS - Use when user needs help with ai-dataset-curator related tasks.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Build text and data classification systems with training and evaluation. TRIGGERS - Use when user needs help with ai-classification-system related tasks.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Runs the CARLA Leaderboard evaluator over a set of routes with your agent — routes/routes-subset/repetitions/track selection, checkpoint and live-results endpoints, resume after a crash, recording, and the debug levels — for leaderboard 1.0, 2.0 or 2.1. Preflights the whole stack (version pairing, maps, ports, agent class, sensor budget) before spending hours, and hands the world back in async mode. Use when the user asks to "run the leaderboard", "evaluate my agent", "run routes_training", "resume an evaluation", or "reproduce my submission locally".
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
The live fieldwork layer for psycho-logic — go look at the actual target market NOW instead of reasoning from your training-data stereotype of it. Use when a psycho-logic move needs grounding in a specific place: the user asks 'what do competitors in [region/country] actually do', 'what's the real price band / norm here', 'how do local leaders position themselves', 'research this market before we localise', 'what signals trust to buyers in X', or is about to launch/adapt in a market you haven't studied. Runs structured web research in the local language — local competitors and their signals, price bands, category conventions, review/trust sources, local terminology, taboos and regulations — and synthesises a compact perception profile the four moves consume. Inspired by Rory Sutherland's *Alchemy*. Routed to from psycho-logic; pair with cultural-context (the lens that tells you what to look for). 日本語・中文の相談（「現地の実態を調べて」「调研当地市场」）にも使う。
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Predicts protein-coding gene structures (exons, introns, UTRs) in eukaryotic genomes with BRAKER3 (RNA-seq + protein evidence), BRAKER1/BRAKER2, GALBA (protein-only), Funannotate (fungi), GeMoMa (homology projection), or Helixer/Tiberius (deep-learning ab initio). Covers the evidence-first tool decision, mandatory soft-masking, the training-set-quality-dominates principle, OrthoDB clade-partition selection, the one-isoform-per-locus and missing-UTR traps, merge/split errors, and reference bias against orphan genes. Use when annotating a newly assembled eukaryotic genome, choosing a gene-prediction pipeline based on available evidence, or diagnosing a poor annotation.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Interactive cell annotation and image QC for IMC/MIBI using napari, napari-imc, Mantis Viewer, and cytomapper, covering the pixels-to-cell-table bridge, overlaying masks to catch segmentation/spillover artifacts, inter-annotator variability as the accuracy ceiling, contrast-as-threshold, and building class-balanced ground-truth label sets. Use when manually labeling cells, generating training data for a classifier, QC-ing segmentation on the image, confirming clusters are spatially real, or choosing an annotation viewer.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Predict peptide-MHC class II (HLA-DR/DQ/DP) binding and presentation for CD4 T-cell epitopes with NetMHCIIpan-4.3 and MixMHC2pred-2.0. Covers why class II is far less reliable than class I (open binding groove, 9-mer register ambiguity, sparse noisy training data, DR>DP>DQ accuracy asymmetry), the DQ/DP heterodimer alpha/beta pairing trap, and the looser 1%/5% %Rank thresholds. Use when predicting CD4 epitopes for vaccine help, mapping class II neoantigens, or scoring long peptides against DR/DQ/DP. For CD8/class I see mhc-binding-prediction.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.