Skill UI

Guides adding PyTorch FSDP2’s fully_shard integration to training loops, covering correct distributed init, bottom-up sharding, mixed precision/offload policies, DTensor-aware optimizers, and distributed checkpointing so oversized models or DeviceMesh sharding goals work reliably.

PyTorch Lightning is a high-level framework built on PyTorch that significantly reduces boilerplate code in deep learning training loops. It automatically handles device management, distributed training (DDP, FSDP, DeepSpeed), mixed precision, and advanced callbacks. It allows researchers to scale models from local machines to supercomputers with minimal code changes, enabling clean, scalable, and best-practice deep learning implementations.

Statsmodels is a comprehensive Python library dedicated to advanced statistical modeling and econometrics. It provides tools for rigorous statistical inference, including Ordinary Least Squares (OLS), Generalized Linear Models (GLM), mixed-effects models, and time series analysis (ARIMA). Use this skill for detailed diagnostics, coefficient estimation, hypothesis testing, and producing publication-ready statistical reports in fields like econometrics and biostatistics.

An interactive workflow for auditing and standardizing naming conventions (tables, columns, measures) in Power BI semantic models (TMDL). It ensures names are human-readable, consistent, and aligned with core business terminology, addressing technical issues like abbreviations, mixed casing, and inappropriate technical prefixes.

This skill provides comprehensive tools for planning studies by calculating the required sample size, minimum detectable effect (MDE), and statistical power. It supports both closed-form statistical tests (t-tests, ANOVA, chi-square) and advanced simulation-based designs (mixed models, logistic/Poisson regression). Use this when planning a study, submitting an IRB protocol, or justifying sample size for a grant.

This guide outlines the rigorous, model-based standards required for data analysis and statistical reporting in Cognitive Psychology manuscripts. It covers essential techniques such as model comparison (using AIC/BIC or Bayes factors), demonstrating parameter and model recovery, implementing mixed or hierarchical Bayesian models, and ensuring all findings are fully reproducible from shared code.

Provides rigorous guidance and structure for statistical analysis in Field Crops Research (FCR). Focuses on implementing linear mixed models that respect complex experimental designs (e.g., split-plot, block, environment). Essential for proper G×E analysis, robust means separation, and accurate evaluation of crop models, ensuring results are methodologically sound and actionable.

This skill provides a comprehensive fit and structural review tool for manuscripts targeting 'Field Crops Research.' It assesses whether an agronomy or crop-physiology paper meets the journal's high standard: rigorous, adequately replicated, multi-environment field experimentation. It guides authors on key topics like resource-use efficiency, crop modeling validation, and required methodological depth (e.g., mixed models, full metadata reporting).

This guide provides a rigorous framework for advanced econometric modeling in quantitative economics. It covers time series, panel data, cointegration, and mixed-frequency techniques (VAR, VECM, GMM, MIDAS). It emphasizes proper model diagnostics, stationarity testing, and grounding models in specific data properties and economic contexts.

A comprehensive guide for auditing and planning the analysis of linguistic manuscripts. It emphasizes that the analysis must credibly support the theoretical claim, covering quantitative modeling (mixed-effects models in R), phonetic measurements, and corpus statistics. Use this skill to stress-test the analysis chain, ensuring rigor and transparency in all generalizations.