Results for “model-merging”

Merges agent or review outputs into one deduplicated, evidence-linked, severity-ranked actionable report.

Train Mixture of Experts (MoE) models using DeepSpeed or HuggingFace, covering architectures, routing, load balancing, and expert parallelism.

Detect poisoned training data and backdoored models across the ML pipeline using statistical analysis, activation clustering, and spectral signatures.

Train Mixture of Experts (MoE) models using DeepSpeed or HuggingFace. Use when training large-scale models with limited compute (5× cost reduction vs dense models), implementing sparse architectures like Mixtral 8x7B or DeepSeek-V3, or scaling model capacity without proportional compute increase. Covers MoE architectures, routing mechanisms, load balancing, expert parallelism, and inference optimization.

Process the Refinery merge queue - collect agent work, detect and resolve conflicts, merge in dependency order, and verify integration.

Guides embedding model migration in Qdrant without downtime, covering alias swap, side-by-side, and hybrid search strategies.

Import object detection models from HuggingFace or NVIDIA NGC into a DeepStream pipeline with automated ONNX download, TensorRT engine build, custom parser, multi-stream benchmark, and PDF report generation.

Integrate a HuggingFace computer vision model into the NVIDIA TAO Toolkit ecosystem, covering the full pipeline from prerequisites to container testing.

Manage machine learning models in the Agent Platform Model Registry: list, describe, upload, update, and delete models and their versions.

Train Mixture of Experts (MoE) models using DeepSpeed or HuggingFace. Use when training large-scale models with limited compute (5× cost reduction vs dense models), implementing sparse architectures like Mixtral 8x7B or DeepSeek-V3, or scaling model capacity without proportional compute increase. Covers MoE architectures, routing mechanisms, load balancing, expert parallelism, and inference optimization.

Evaluates text-to-image models on spatial relationship accuracy using the VISOR metric, separating object detection from spatial correctness to reveal biases like object priority and merging.

Entrena modelos de machine learning con Scikit-learn, LightGBM y XGBoost, desde un baseline hasta un modelo productivo con validación robusta y explicabilidad.

DEPRECATED in v5.6.0 — see joint-multi-tf-v560 skill. Documents the v5.2.0 dual-model approach (train separate 15Min/1Hour models, combine via weighted voting). Still relevant for: (1) loading legacy v5.5.0 dual models, (2) understanding the historical aggregation layer, (3) resampling pattern via origin='start'.

Turns model work into a production ML system with data contracts, reproducible training, quality gates, deployable artifacts, and monitoring.

Orchestrates three frontier models to debate a question and synthesizes their best insights into a single superior answer.

Model, simulate, and optimize Serializer/Deserializer (SerDes) systems — serial and parallel links — using MATLAB SerDes Toolbox. Design NRZ and PAM-N links (PAM3 through PAM16) — explore equalization architectures (FFE, CTLE, DFE), sweep or optimize parameters with genetic algorithms, and characterize channels from loss models, S-parameter files, or crosstalk scenarios. Process captured waveforms through equalization chains, build eye diagrams, and decompose jitter. Deliver IBIS-AMI models for Tx, Rx, Redriver, or Retimer by exporting to Simulink and compiling .ami/.ibs/.dll/.so files. Covers the full arc from initial design exploration and parameter optimization to compliance testing and compiled model validation, including custom datapath blocks for nonstandard equalization.

Turn model work into a production ML system with data contracts, reproducible training, quality gates, deployable artifacts, and monitoring.

Model a leveraged buyout end to end — sources and uses, tranche-level debt with amortization and cash sweep, levered free cash flow, and sponsor returns (IRR and MOIC) with returns attribution and credit-stat tracking.

The layers trade timeliness against definitiveness.

Guides machine learning development with experiment tracking, hyperparameter optimization, model registry, and MLOps pipeline integration.

Selects the right Gemma model for a task, recommends deployment tooling (Gradio, Transformers.js, Vertex AI, MLX), and applies optimizations like MTP and QAT.

Production machine-learning engineering workflow for data contracts, reproducible training, model evaluation, deployment, monitoring, and rollback. Use when building, reviewing, or hardening ML systems beyond one-off notebooks.

Integration of BDI agent architecture with soft systems methodology for complex organizational problem-solving

A model in production is never just weights.

Model-trader version compatibility protocol: Embed version metadata in checkpoints, validate at load time. Trigger when: (1) training and live trading versions diverge, (2) models fail to load, (3) action interpretation issues.

Compress large language models by 40-60% with minimal accuracy loss using one-shot pruning techniques like Wanda and SparseGPT, enabling faster inference and deployment on constrained hardware.

Train machine learning models end-to-end, covering data loading, preprocessing, architecture selection, training loops, validation, and checkpointing. Use when the user requests model training or provides relevant inputs for this workflow.