https://uaipedro.r-universe.devRecent package updates in uaipedroR-universehttps://github.com/uaipedro.pnghttps://uaipedro.r-universe.devFri, 09 Oct 2026 16:23:25 GMTpedromambelli@gmail.com (Pedro Mambelli Fernandes)Supervised learning nodes for trama workflows, from linear baselines and interpretable CART and FIGS models to random forests, support vector machines and gradient boosted trees. Provides reproducible training and test splits, prediction, evaluation and model inspection. Learning engines are optional and loaded only for the selected model. Neural networks are deliberately outside the scope of this collection.https://github.com/r-universe/uaipedro/actions/runs/38094025472Fri, 09 Oct 2026 16:23:25 GMTtrama.ml0.5.8successhttps://uaipedro.r-universe.devhttps://github.com/uaipedro/tramapedromambelli@gmail.com (Pedro Mambelli Fernandes)Build and run computation graphs interactively. Nodes are typed R functions with declared ports and parameters; a flow is a plain JSON document; execution is pull-based, content-addressed and incremental, so editing a parameter recomputes only what depends on it. Every node renders its own result in the canvas with a registered renderer, making each step of a pipeline visible rather than implied. Domain knowledge lives entirely in registrable collections, never in the core, so new domains are added without modifying the package.https://github.com/r-universe/uaipedro/actions/runs/38092775140Fri, 09 Oct 2026 16:23:25 GMTtrama0.5.16successhttps://uaipedro.r-universe.devhttps://github.com/uaipedro/tramapedromambelli@gmail.com (Pedro Mambelli Fernandes)Nodes for planning experiments in a trama graph: a declarative design node that builds and randomizes completely randomized, randomized block, Latin square, factorial (with or without confounding in blocks), fractional factorial 2^(k-p), central composite, split-plot, strip-plot, balanced incomplete block, repeated measures, crossover (Williams) and multi-site designs, with every factor's role, unit of assignment and randomization scope made explicit. The plan carries its metadata and becomes a plain table for the analysis nodes of trama.models. Builds on trama.data for tables, trama.view for plots and trama.models for the fitted models that the analysis nodes read.https://github.com/r-universe/uaipedro/actions/runs/38094023017Fri, 09 Oct 2026 16:23:25 GMTtrama.experiments0.2.8successhttps://uaipedro.r-universe.devhttps://github.com/uaipedro/tramapedromambelli@gmail.com (Pedro Mambelli Fernandes)Nodes for multivariate analysis in a trama graph: principal component analysis with scree, biplot and correlation circle; factor analysis by maximum likelihood or principal axis, with orthogonal (varimax, quartimax, equamax) and oblique (promax, oblimin) rotations; linear and quadratic discriminant analysis with exact leave-one-out predictions; logistic regression (binary and multinomial) as a classifier; jackknife bias, standard errors and pseudovalues for component, factor, discriminant and logistic statistics; and the diagnostics that come first (KMO, Bartlett's sphericity, Horn's parallel analysis, Box's M); distance matrices (Euclidean, Mahalanobis D2, Gower), hierarchical clustering with cophenetic correlation, k-means and Tocher's optimization method; and one-way MANOVA with optional blocks. Ships example data with known structure. Builds on trama.data for tables, trama.view for plots and trama.models, whose model contract the classifiers implement (prediction, confusion and ROC are its nodes).https://github.com/r-universe/uaipedro/actions/runs/38094027603Fri, 09 Oct 2026 16:23:25 GMTtrama.multi0.3.8successhttps://uaipedro.r-universe.devhttps://github.com/uaipedro/tramapedromambelli@gmail.com (Pedro Mambelli Fernandes)Nodes that draw ggplot2 graphics inside a trama graph. Each node renders one HD image at a user-chosen aspect ratio.https://github.com/r-universe/uaipedro/actions/runs/38092781469Fri, 09 Oct 2026 16:23:25 GMTtrama.view0.2.8successhttps://uaipedro.r-universe.devhttps://github.com/uaipedro/tramapedromambelli@gmail.com (Pedro Mambelli Fernandes)Nodes for time series in a trama graph: operators (differencing, lags, transforms, windows, moving averages, aggregation), classical and STL decomposition, ARIMA, ETS and Holt-Winters models with forecasts and benchmarks, stationarity and white-noise tests, and series-specific plots such as correlograms, seasonal and lag plots; multivariate series with VAR and VECM models, Granger causality, Engle-Granger and Johansen cointegration tests, impulse responses and residual diagnostics. Builds on trama.data for tables and on trama.view for plots, without duplicating either.https://github.com/r-universe/uaipedro/actions/runs/38094029654Fri, 09 Oct 2026 16:23:25 GMTtrama.series0.9.1successhttps://uaipedro.r-universe.devhttps://github.com/uaipedro/tramapedromambelli@gmail.com (Pedro Mambelli Fernandes)Nodes for survey sampling in a trama graph: sample size for means and proportions with finite population correction, design effect and non-response; stratified sample size with proportional, Neyman, cost optimal and equal allocation; cluster sample size from the intraclass correlation; selection of simple random, systematic, stratified, probability proportional to size, one-stage and two-stage cluster samples; declared designs and post-stratification; design-based estimation of means, totals, proportions and ratios, in domains, with the ultimate cluster variance estimator; and Monte Carlo simulation of a design against a known population. The selected sample carries its design, so estimation never asks for it again. Builds on trama.data for tables and trama.view for plots.https://github.com/r-universe/uaipedro/actions/runs/38092796613Fri, 09 Oct 2026 16:23:25 GMTtrama.sampling0.3.8successhttps://uaipedro.r-universe.devhttps://github.com/uaipedro/tramapedromambelli@gmail.com (Pedro Mambelli Fernandes)Nodes for loading, reshaping and summarising tabular data in a trama graph. Thin wrappers over readr, dplyr and tidyr.https://github.com/r-universe/uaipedro/actions/runs/38560124357Fri, 09 Oct 2026 16:23:25 GMTtrama.data0.4.4successhttps://uaipedro.r-universe.devhttps://github.com/uaipedro/tramapedromambelli@gmail.com (Pedro Mambelli Fernandes)Nodes for geostatistics in a trama graph: declaring a regionalised variable with coordinates, projection and boundary; exploratory spatial display; classical and robust empirical variograms, omnidirectional or directional, with large-scale trend removed; fitting spherical, exponential, Gaussian and Matern models by ordinary and weighted least squares, reporting nugget, partial sill, range and practical range; and ordinary and simple kriging onto a boundary-clipped grid, carrying the kriging variance so the prediction map always has an uncertainty map beside it. The point object carries its coordinates, projection and boundary, so no later node asks for them again. Builds on trama.data for tables and trama.view for plots.https://github.com/r-universe/uaipedro/actions/runs/38092805763Fri, 09 Oct 2026 16:23:25 GMTtrama.spatial0.1.1successhttps://uaipedro.r-universe.devhttps://github.com/uaipedro/tramapedromambelli@gmail.com (Pedro Mambelli Fernandes)Nodes for statistical models in a trama graph: linear and generalized linear models from columns or a typed formula; analysis of variance for the classical experimental designs (completely randomized, randomized blocks, latin square, factorial and split plot) with type I, II and III sums of squares; linear mixed models with lme4 and lmerTest; estimated marginal means, compact letter displays and pairwise or Dunnett comparisons with emmeans; Duncan, Waller-Duncan and Scott-Knott grouping; effect sizes and coefficient forest plots; residual diagnostics and assumption tests; and the classical hypothesis tests with the Dunn post hoc. Every p-value is shown on a logarithmic gauge with significance stars. Builds on trama.data for tables and trama.view for plots.https://github.com/r-universe/uaipedro/actions/runs/38092791215Fri, 09 Oct 2026 16:23:25 GMTtrama.models0.6.9successhttps://uaipedro.r-universe.devhttps://github.com/uaipedro/tramapedromambelli@gmail.com (Pedro Mambelli Fernandes)Motor de instalação e atualização do trama: lê o manifesto de release, calcula o status da instalação local, instala releases numa biblioteca isolada (só pacotes binários) e serve a tela de início em Shiny, de onde o editor é aberto.https://github.com/r-universe/uaipedro/actions/runs/38092809105Fri, 09 Oct 2026 16:23:25 GMTtrama.launcher0.2.2successhttps://uaipedro.r-universe.devhttps://github.com/uaipedro/tramapedromambelli@gmail.com (Pedro Mambelli Fernandes)Query local files and databases with SQL and return tables to trama flows.https://github.com/r-universe/uaipedro/actions/runs/38092802872Fri, 09 Oct 2026 16:23:25 GMTtrama.sql0.1.0successhttps://uaipedro.r-universe.devhttps://github.com/uaipedro/trama