Results for “pbkdf2”

Adds PyTorch FSDP2 (fully_shard) to training scripts with correct init, sharding, mixed precision/offload config, and distributed checkpointing. Use when models exceed single-GPU memory or when you need DTensor-based sharding with DeviceMesh.

Adds PyTorch FSDP2 (fully_shard) to training scripts with correct init, sharding, mixed precision/offload config, and distributed checkpointing. Use when models exceed single-GPU memory or when you need DTensor-based sharding with DeviceMesh.

Differential gene expression analysis (Python DESeq2). Identify DE genes from bulk RNA-seq counts, Wald tests, FDR correction, volcano/MA plots, for RNA-seq analysis.

Kernel-Mode Driver Framework (KMDF), the Microsoft-recommended way to write Windows kernel-mode drivers. Covers DriverEntry, EvtDeviceAdd, IRPs and IOCTLs, I/O queues, PnP and Power state machines, IRQL discipline, memory pools, WPP tracing, SAL annotations, and Driver Verifier. USE WHEN: user mentions "KMDF", "WDF kernel", "Windows kernel driver", "DriverEntry", "WdfDriverCreate", "EvtDeviceAdd", "IRP", "IOCTL", "DISPATCH_LEVEL", "PASSIVE_LEVEL", "NTSTATUS", "PoolTag", "WdfRequestComplete" DO NOT USE FOR: UMDF v2 (use `wdf-umdf`), classic WDM-only drivers, file-system filters (FltMgr is a separate framework)

Generate images with Flux 2 Klein (Black Forest Labs' distilled fast variant of Flux 2) on RunComfy — bundled with the model's documented prompting patterns so the skill gets sharper output than naive prompting against the same model. Documents Flux 2 Klein's strengths (sub-second latency, multi-reference brand styling, declarative subject-first prompts), the step-count strategy (4–8 for fast iteration, ~25 for polish), the 9B vs 4B variant trade-off, and when to route to Flux 2 Pro / Seedream 5 / GPT Image 2 instead. Calls `runcomfy run blackforestlabs/flux-2-klein/9b/text-to-image` (or `/4b/`) through the local RunComfy CLI. Triggers on "flux 2 klein", "flux-2-klein", "flux klein", "BFL flux 2", or any explicit ask to generate with this model.

Parameter-efficient fine-tuning for LLMs using LoRA, QLoRA, and 25+ methods. Use when fine-tuning large models (7B-70B) with limited GPU memory, when you need to train <1% of parameters with minimal accuracy loss, or for multi-adapter serving. HuggingFace's official library integrated with transformers ecosystem.

Diffusion-based molecular docking. Predict protein-ligand binding poses from PDB/SMILES, confidence scores, virtual screening, for structure-based drug design. Not for affinity prediction.

Expert en Obsidian avancé (vaults, plugins, dataview, templates, graph view, PKM workflows, sync, publish)

Differential gene expression analysis (Python DESeq2). Identify DE genes from bulk RNA-seq counts, Wald tests, FDR correction, volcano/MA plots, for RNA-seq analysis.

Fine-tune large language models by training less than 1% of parameters using LoRA, QLoRA, and 25+ adapter methods, enabling efficient adaptation on limited GPU memory.

User-Mode Driver Framework v2 (UMDF). User-mode driver model that uses the same WDF object model as KMDF but runs in a host process (WUDFHost.exe) protected by the reflector. Required for some categories (Indirect Display Drivers, many sensor and camera drivers) and recommended for any driver that doesn't strictly need kernel mode. USE WHEN: user mentions "UMDF", "WUDFHost", "user-mode driver", "reflector", "IDD", "ISensor", "WDFHOST", "UMDF v2", "FX2" DO NOT USE FOR: KMDF (use `wdf-kmdf`), classic UMDF v1 (deprecated, COM-based)

Adds PyTorch FSDP2 (fully_shard) to training scripts with correct init, sharding, mixed precision/offload config, and distributed checkpointing. Use when models exceed single-GPU memory or when you need DTensor-based sharding with DeviceMesh.

Create Pydantic models following the multi-model pattern with Base, Create, Update, Response, and InDB variants for clean API contracts in Python applications using Pydantic v2.

Diagnose and optimize slow-loading, high-memory, or low-FPS USD scenes using a structured workflow with profiling, validation, and mutation phases.

Run differential gene expression analysis on bulk RNA-seq count matrices with PyDESeq2, the Python port of DESeq2 — size-factor normalization, dispersion estimation, Wald tests, FDR (Benjamini-Hochberg) correction, and volcano/MA plots. Use when identifying differentially expressed genes between conditions from raw bulk RNA-seq counts. Part of the AlterLab Academic Skills suite.

Idefics2: An 8B Parameters Multimodal Model

Use when a GFF file needs bcftools/csq-compatible gene and transcript attributes before consequence annotation.

Perform differential gene expression analysis for bulk RNA-seq data using PyDESeq2, supporting formulaic designs, Wald tests, FDR correction, LFC shrinkage, and result visualization.

Configure secure OAuth 2.0 authorization flows including Authorization Code with PKCE, Client Credentials, and Device Authorization Grant, covering flow selection, PKCE implementation, token lifecycle management, scope design, and alignment with OAuth 2.1 security requirements.

Parameter-efficient fine-tuning for LLMs using LoRA, QLoRA, and 25+ methods. Use when fine-tuning large models (7B-70B) with limited GPU memory, when you need to train <1% of parameters with minimal accuracy loss, or for multi-adapter serving. HuggingFace's official library integrated with transformers ecosystem.

Generate images with Flux 2 Klein (Black Forest Labs' distilled fast variant of Flux 2) on RunComfy — bundled with the model's documented prompting patterns so the skill gets sharper output than naive prompting against the same model. Documents Flux 2 Klein's strengths (sub-second latency, multi-reference brand styling, declarative subject-first prompts), the step-count strategy (4–8 for fast iteration, ~25 for polish), the 9B vs 4B variant trade-off, and when to route to Flux 2 Pro / Seedream 5 / GPT Image 2 instead. Calls `runcomfy run blackforestlabs/flux-2-klein/9b/text-to-image` (or `/4b/`) through the local RunComfy CLI. Triggers on "flux 2 klein", "flux-2-klein", "flux klein", "BFL flux 2", or any explicit ask to generate with this model.

Parameter-efficient fine-tuning for LLMs using LoRA, QLoRA, and 25+ methods. Use when fine-tuning large models (7B-70B) with limited GPU memory, when you need to train <1% of parameters with minimal accuracy loss, or for multi-adapter serving. HuggingFace's official library integrated with transformers ecosystem.

Implements FIDO2/WebAuthn hardware security key authentication with registration, authentication, YubiKey enrollment, and passkey migration using the python-fido2 library.

Generate images with Flux 2 Klein (Black Forest Labs' distilled fast variant of Flux 2) on RunComfy — bundled with the model's documented prompting patterns so the skill gets sharper output than naive prompting against the same model. Documents Flux 2 Klein's strengths (sub-second latency, multi-reference brand styling, declarative subject-first prompts), the step-count strategy (4–8 for fast iteration, ~25 for polish), the 9B vs 4B variant trade-off, and when to route to Flux 2 Pro / Seedream 5 / GPT Image 2 instead. Calls `runcomfy run blackforestlabs/flux-2-klein/9b/text-to-image` (or `/4b/`) through the local RunComfy CLI. Triggers on "flux 2 klein", "flux-2-klein", "flux klein", "BFL flux 2", or any explicit ask to generate with this model.

Train reinforcement learning agents at millions of steps per second using optimized PPO, vectorized environments, and multi-agent support.

Diffusion-based molecular docking. Predict protein-ligand binding poses from PDB/SMILES, confidence scores, virtual screening, for structure-based drug design. Not for affinity prediction.

Generate image captions, answer visual questions, and perform image-text retrieval using BLIP-2's Q-Former architecture with frozen vision encoders and LLMs.

Imported skill reference from anthropic

Diffusion-based molecular docking. Predict protein-ligand binding poses from PDB/SMILES, confidence scores, virtual screening, for structure-based drug design. Not for affinity prediction.

Use when designing a Digital Down Converter (DDC) using dsphdl System objects. Triggers on requests involving DDC design, frequency down-conversion for FPGA/ASIC, NCO + mixer + decimation filter chains, fractional/non-integer sample rate conversion, or HDL-optimized receiver front-end signal processing.

Inspects and debugs OAuth 2.0 authorization flows including PKCE, client credentials, and device code grants, validating tokens, decoding claims, and tracing redirect chains.

Generate typed SDK bindings (Pydantic v2 models, TypeScript types) from the Bloodbank JSON Schema tree at `bloodbank/schemas/bloodbank/v1/**` (Draft 2020-12). Uses `datamodel-code-generator` and `json-schema-to-typescript` — NOT the deprecated hand-rolled Holyfields generators. Use when generating event-contract bindings for a Bloodbank consumer, importing typed CloudEvents envelopes into another project, or scaffolding a `bloodbank-contracts` SDK package. Trigger keywords — "bloodbank SDK", "bloodbank contracts", "Pydantic from bloodbank schemas", "TypeScript types for bloodbank events", "bloodbank.v1.* types", "event contract bindings", "datamodel-code-generator", "json-schema-to-typescript", "regenerate SDK", "CloudEvents bindings". Do NOT use for schema authoring (edit `bloodbank/schemas/` directly per `docs/event-naming.md` §12), runtime envelope validation (use `BLOODBANK_HOOK_VALIDATE=1`), schema-tree consistency (`mise run smoketest:schemas`), or generic JSON Schema codegen unrelated to Bloodbank.

Parameter-efficient fine-tuning for LLMs using LoRA, QLoRA, and 25+ methods. Use when fine-tuning large models (7B-70B) with limited GPU memory, when you need to train <1% of parameters with minimal accuracy loss, or for multi-adapter serving. HuggingFace's official library integrated with transformers ecosystem.

Use when converting GenBank format files to table format as part of the Entrez Direct toolkit from bioconda.

Vision-language pre-training framework bridging frozen image encoders and LLMs. Use when you need image captioning, visual question answering, image-text retrieval, or multimodal chat with state-of-the-art zero-shot performance.