diff --git a/.github/workflows/ci.yml b/.github/workflows/ci.yml index 5d7b6ae..358040c 100644 --- a/.github/workflows/ci.yml +++ b/.github/workflows/ci.yml @@ -32,12 +32,21 @@ jobs: - name: every member organisation and repository is on it run: | rc=0 - for org in go-ndarray go-fft go-numerics go-statistics go-dataframe go-plot go-notebook; do + for org in go-ndarray go-fft go-numerics go-statistics go-dataframe go-plot go-notebook \ + go-bignum go-sparse go-approx go-dsp; do for url in "https://$org.github.io/" "https://$org.github.io/docs/"; do grep -qF "$url" public/index.html \ || { echo "::error::$url is not on the page"; rc=1; } done done + # The planned organisations have no site yet: their card links to + # GitHub, the organisation itself (minified HTML drops the quotes, + # so the link ends at '>' or '"', not at '/'). + for org in go-symbolic go-mlkit go-autodiff go-geometry go-fem \ + go-graphalgo go-control go-units go-sciio; do + grep -qE "https://github.com/$org[\">]" public/index.html \ + || { echo "::error::https://github.com/$org is not on the page"; rc=1; } + done for repo in go-scikit.github.io brand .github; do grep -qF "go-scikit/$repo" public/index.html \ || { echo "::error::go-scikit/$repo is not on the page"; rc=1; } diff --git a/hugo.toml b/hugo.toml index 5de8033..bf6fd3a 100644 --- a/hugo.toml +++ b/hugo.toml @@ -22,16 +22,26 @@ disableKinds = ['taxonomy', 'term', 'rss', 'sitemap', '404'] [params] description = "go-scikit: the scientific stack for Go, in pure Go (CGO=0). N-dimensional arrays and NumPy's random numbers, FFTs, optimisation, integration and special functions, probability and statistics, labelled tables, figures and compiled, reactive notebooks, each measured against its Python reference." -# One entry per member organisation. The cards in #orgs, the strip under the -# hero, the stack diagram in #stack and the ci check that the page names -# every organisation all come from this list. `layer` places it in the stack: -# 0 the arrays, 1 the libraries that compute on them, 2 the notebook. -# Every statement is from the organisation's own profile, README or landing. +# One entry per member organisation, the established ones and those of the +# parity plan alike: the cards in #orgs, the strip under the hero, the stack +# diagram in #stack and the ci check that the page names every organisation +# all come from this list, so every organisation is shown the same way. +# `layer` places it in the stack: 0 the arrays, 1 the libraries that compute +# on them, 2 the notebook. `state` is the release status (the highest tag of +# `lib`, checked with the tags API). `live` means the organisation's landing +# and documentation are published (otherwise the card links to GitHub, not to +# pages that do not exist); `brand` means its logo is published (otherwise +# the card shows the go-scikit mark). Every statement is from the +# organisation's own profile, README or landing, or from the roadmap's +# PLAN.md and STATUS.md. [[params.orgs]] name = "go-ndarray" layer = 0 short = "arrays, numpy.linalg, numpy.random" lib = "ndarray" + state = "ndarray v0.18.0" + live = true + brand = true blurb = "N-dimensional arrays in 13 dtypes with NumPy 2 promotion, broadcasting, strided views and reductions, multicore and SIMD. Package linalg is numpy.linalg, fft is numpy.fft, and random is numpy.random with NumPy's numbers for the same seed." playground = "https://go-ndarray.github.io/playground/" [[params.orgs]] @@ -39,6 +49,9 @@ disableKinds = ['taxonomy', 'term', 'rss', 'sitemap', '404'] layer = 0 short = "numpy.fft, scipy.fft" lib = "fft" + state = "fft v0.27.3" + live = true + brand = true blurb = "The numpy.fft / scipy.fft equivalent: complex and real transforms of any length, N-D, the DCT and DST, single precision, windows, with no dependency on FFTW. AVX2 and AVX-512 kernels on amd64, NEON on arm64, bit-identical to the Go code." playground = "https://go-fft.github.io/playground/" [[params.orgs]] @@ -46,6 +59,9 @@ disableKinds = ['taxonomy', 'term', 'rss', 'sitemap', '404'] layer = 1 short = "optimize, integrate, special" lib = "optimize" + state = "optimize, integrate, special v0.2.0" + live = true + brand = true blurb = "Numerical methods ported statement by statement from their reference algorithms: minimisation, roots, least squares and linear programming (optimize); QUADPACK quadrature and ODE solvers (integrate); gamma, beta, error, Bessel and zeta functions measured against mpmath (special)." playground = "https://go-numerics.github.io/playground/" [[params.orgs]] @@ -53,6 +69,9 @@ disableKinds = ['taxonomy', 'term', 'rss', 'sitemap', '404'] layer = 1 short = "distributions, statistics, tests" lib = "stats" + state = "stats v0.1.0" + live = true + brand = true blurb = "32 univariate and 3 multivariate distributions with every method of SciPy's frozen distributions, descriptive statistics with NumPy's 13 quantile methods, and the statistical tests, from t-tests to Shapiro–Wilk and Anderson–Darling." playground = "https://go-statistics.github.io/playground/" [[params.orgs]] @@ -60,19 +79,118 @@ disableKinds = ['taxonomy', 'term', 'rss', 'sitemap', '404'] layer = 1 short = "labelled tables" lib = "dataframe" + state = "dataframe v0.2.0" + live = true + brand = true blurb = "Series, DataFrame and Index whose numeric columns are go-ndarray arrays: CSV, Arrow and Parquet, group-by, joins, reshaping and time series, giving pandas' answers." [[params.orgs]] name = "go-plot" layer = 1 short = "figures, SVG and PNG" lib = "plot" + state = "plot v0.1.1" + live = true + brand = true blurb = "matplotlib's object model and defaults, figures to SVG and PNG, real TeX in the labels; ticks, colormaps and layout tested equal to matplotlib where they can be equal." [[params.orgs]] name = "go-notebook" layer = 2 short = "compiled, reactive notebooks" lib = "notebook" + state = "notebook v0.2.0" + live = true + brand = true blurb = "A notebook is a Markdown file whose go blocks are cells. The kernel compiles it into a real Go program, orders the cells by what they use, and runs only what changed, keeping values across runs." +[[params.orgs]] + name = "go-bignum" + layer = 1 + short = "arbitrary precision, oracle harness" + lib = "bigfloat" + state = "bigfloat in development, not yet tagged" + live = true + brand = true + blurb = "Correctly rounded functions on math/big's big.Float (exponentials, logarithms, roots, trigonometric, hyperbolic and power functions, constants), checked against mpmath, and the Go oracle harness the other organisations validate with. Ball arithmetic and number theory are planned." +[[params.orgs]] + name = "go-sparse" + layer = 1 + short = "scipy.sparse, iterative solvers" + lib = "sparse" + state = "sparse v0.1.0" + live = true + brand = true + blurb = "CSR, CSC and COO matrices with int32 indices as in scipy.sparse, fill-reducing orderings (AMD, COLAMD, RCM) and preconditioned CG, GMRES, MINRES and LSQR with SciPy's iteration counts; every measured case at or below SciPy's time. Direct solvers and sparse eigensolvers are planned." +[[params.orgs]] + name = "go-approx" + layer = 1 + short = "interpolation, splines, polynomials" + lib = "interp" + state = "interp v0.2.0, poly v0.1.0" + live = true + brand = true + blurb = "interp is numpy.interp, interp1d, PCHIP, Akima, cubic, B- and smoothing splines and gridded interpolation; poly is numpy.polyfit, roots and the five numpy.polynomial bases, at or below NumPy's time. Chebyshev and AAA approximation and scattered data are planned." +[[params.orgs]] + name = "go-dsp" + layer = 1 + short = "scipy.signal" + lib = "signal" + state = "signal v0.1.0" + live = true + brand = true + blurb = "scipy.signal in Go: IIR and FIR design (Butterworth to elliptic, Remez, least squares), filtering, convolution, Welch, spectrograms, resampling and peak finding, all 28 measured cases at or below SciPy's time. Wavelets and audio I/O are planned." +[[params.orgs]] + name = "go-symbolic" + layer = 1 + short = "computer algebra" + state = "planned (wave 4)" + blurb = "Computer algebra: expressions, simplification, differentiation, integration and equation solving, the core of what Mathematica does symbolically." +[[params.orgs]] + name = "go-mlkit" + layer = 1 + short = "machine learning" + state = "planned (wave 3)" + blurb = "Machine learning on go-ndarray arrays: clustering, classification and regression models, and model selection." +[[params.orgs]] + name = "go-autodiff" + layer = 1 + short = "automatic differentiation" + state = "planned (wave 2)" + blurb = "Automatic differentiation, forward and reverse mode, for the optimisers, solvers and models of the other organisations." +[[params.orgs]] + name = "go-geometry" + layer = 1 + short = "computational geometry" + state = "planned (wave 2)" + blurb = "Robust predicates, convex hulls, Delaunay triangulations, Voronoi diagrams, k-d trees and mesh generation." +[[params.orgs]] + name = "go-fem" + layer = 1 + short = "finite elements, PDEs" + state = "planned (wave 3)" + blurb = "Finite elements and PDE solvers on the meshes of go-geometry and the sparse solvers of go-sparse." +[[params.orgs]] + name = "go-graphalgo" + layer = 1 + short = "graph algorithms" + state = "planned (wave 3)" + blurb = "Graph algorithms: traversals, shortest paths, flows, matchings, connectivity and centrality." +[[params.orgs]] + name = "go-control" + layer = 1 + short = "control systems, DAEs" + state = "planned (wave 3)" + blurb = "Control systems (transfer functions, state space, analysis and design) and DAE solvers." +[[params.orgs]] + name = "go-units" + layer = 1 + short = "units, physical constants" + state = "planned (wave 2)" + blurb = "Units of measure with dimensional checking, and the physical constants." +[[params.orgs]] + name = "go-sciio" + layer = 0 + short = "scientific file formats" + state = "planned (wave 2)" + blurb = "Scientific file formats read into and written from go-ndarray arrays: HDF5, MAT, NetCDF, FITS, npy, Matrix Market, WAV and FLAC." # The cards of #measured and their count come from this list. Each statement # is from the document its link names. @@ -143,67 +261,6 @@ disableKinds = ['taxonomy', 'term', 'rss', 'sitemap', '404'] what = "Ball arithmetic, N-digit special functions and number theory; computer algebra." homes = "go-bignum, go-symbolic" -[[params.neworgs]] - name = "go-bignum" - state = "bigfloat in development, not yet tagged" - what = "arbitrary precision, number theory, the Go oracle harness" - site = "https://go-bignum.github.io/" - docs = "https://go-bignum.github.io/docs/" -[[params.neworgs]] - name = "go-sparse" - state = "sparse v0.1.0 released" - what = "sparse matrices, orderings, iterative and direct solvers, sparse eigensolvers" - site = "https://go-sparse.github.io/" - docs = "https://go-sparse.github.io/docs/" -[[params.neworgs]] - name = "go-approx" - state = "interp and poly v0.1.0 released" - what = "interpolation, splines, polynomials, Chebyshev and AAA approximation, scattered data" - site = "https://go-approx.github.io/" - docs = "https://go-approx.github.io/docs/" -[[params.neworgs]] - name = "go-dsp" - state = "signal v0.1.0 released" - what = "filter design and filtering, spectral estimation, resampling, wavelets, audio I/O" - site = "https://go-dsp.github.io/" - docs = "https://go-dsp.github.io/docs/" -[[params.neworgs]] - name = "go-symbolic" - state = "planned" - what = "computer algebra" -[[params.neworgs]] - name = "go-mlkit" - state = "planned" - what = "machine learning, clustering, classification, model selection" -[[params.neworgs]] - name = "go-autodiff" - state = "planned" - what = "automatic differentiation" -[[params.neworgs]] - name = "go-geometry" - state = "planned" - what = "robust predicates, convex hull, Delaunay, Voronoi, k-d trees, mesh generation" -[[params.neworgs]] - name = "go-fem" - state = "planned" - what = "finite elements and PDE solvers" -[[params.neworgs]] - name = "go-graphalgo" - state = "planned" - what = "graph algorithms" -[[params.neworgs]] - name = "go-control" - state = "planned" - what = "control systems and DAE solvers" -[[params.neworgs]] - name = "go-units" - state = "planned" - what = "units and physical constants" -[[params.neworgs]] - name = "go-sciio" - state = "planned" - what = "scientific file formats (HDF5, MAT, NetCDF, FITS, npy, Matrix Market, WAV/FLAC)" - # Every field below is rendered by layouts/index.html (#repos). A field # nothing renders does not stay true. [[params.repos]] diff --git a/layouts/index.html b/layouts/index.html index 8f3a0e9..6c1416b 100644 --- a/layouts/index.html +++ b/layouts/index.html @@ -31,7 +31,7 @@

Science in Go,
{{/* The sheet's .accent paints with --grad, base to deep: on a dark page this logo's dark slate stops would vanish. --accent and --accent-deep are the same two in light and the bright stop in dark. */}}all the way down.

-

go-scikit is the umbrella over {{ len $orgs }} organisations that together cover what a Python notebook reaches for in NumPy, SciPy, pandas and matplotlib: arrays, FFTs, numerical methods, statistics, labelled tables, figures and notebooks. Pure Go, CGO=0, and each library measured against its Python reference, with the documents that say where it is slower or less accurate.

+

go-scikit is the umbrella over {{ len $orgs }} organisations that together cover what a Python notebook reaches for in NumPy, SciPy, pandas and matplotlib: arrays, FFTs, numerical methods, statistics, labelled tables, figures and notebooks, and, on the way to parity with MATLAB and Mathematica, sparse matrices, interpolation, signal processing and arbitrary precision. Pure Go, CGO=0, and each library measured against its reference, with the documents that say where it is slower or less accurate.

Write a first notebook → {{ partial "icons/book.html" . }}The organisations @@ -48,7 +48,7 @@

Science in Go,
{{/* The sheet's .accent paints with --grad,
{{ len $orgs }} organisations — one stack, each with its own site and documentation -
{{ range $orgs }}{{ .name }}{{ end }}
+
{{ range $orgs }}{{ .name }}{{ end }}
{{/* The diagram is drawn from params.orgs: layer 2 on top, layer 0 at the @@ -61,8 +61,8 @@

Arrays at the bottom, notebooks on top.

{{ range $layer := slice 2 1 0 }}
{{ range where $orgs "layer" $layer }} - - {{ .name }} logo + + {{ partial "org-logo.html" (dict "org" . "size" 36 "radius" 9) }} {{ .name }}
{{ .short }}
{{ end }} @@ -73,15 +73,16 @@

Arrays at the bottom, notebooks on top.

The organisations -

{{ len $orgs }} organisations, one site each.

-

Each organisation has its own landing, documentation and release cycle. The libraries are in the organisations; this page only points at them.

+

{{ len $orgs }} organisations, one stack.

+

Each organisation has its own landing, documentation and release cycle; each card gives its release status. The libraries are in the organisations; this page only points at them. Planned organisations link to GitHub and to the plan until they have a site.

{{ range $orgs }}
- {{ .name }} logo +
{{ partial "org-logo.html" (dict "org" . "size" 42 "radius" 12) }}

{{ .name }}

+

{{ .state }}

{{ .blurb | safeHTML }}

-

Site →   Docs →   Source →{{ with .playground }}   Playground →{{ end }}

+

{{ if .live }}Site →   Docs →{{ if .lib }}   Source →{{ end }}{{ with .playground }}   Playground →{{ end }}{{ else }}GitHub →   Plan →{{ end }}

{{ end }}
@@ -101,18 +102,13 @@

{{ len .Site.Params.claims }} claims, each with its evidence.

Roadmap

Parity with MATLAB and Mathematica, in four waves.

-

The goal is feature parity with MATLAB (base and toolboxes) and Mathematica, in pure Go (CGO=0), each piece validated against the reference tools of its domain. The plan, the home of each domain and the order of the work are in the roadmap repository: PLAN.md, STATUS.md and the bibliography. Four packages of the new organisations have been released at v0.1.0: go-approx/interp (speed comparison in its docs/perf.md), go-approx/poly, go-dsp/signal and go-sparse/sparse; go-bignum/bigfloat is in development and not yet tagged. The vector-update kernels Axpy, Xpby, AxpyDot and Scale shipped in go-simd/floats v0.4.0.

+

The goal is feature parity with MATLAB (base and toolboxes) and Mathematica, in pure Go (CGO=0), each piece validated against the reference tools of its domain. The plan, the home of each domain and the order of the work are in the roadmap repository: PLAN.md, STATUS.md and the bibliography. Four packages of the new organisations have been released: go-approx/interp v0.2.0 (speed comparison in its docs/perf.md), go-approx/poly, go-dsp/signal and go-sparse/sparse v0.1.0, each at or below its reference's time on every case measured; go-bignum/bigfloat is in development and not yet tagged. The vector-update kernels Axpy, Xpby, AxpyDot and Scale shipped in go-simd/floats v0.4.0.

{{ range .Site.Params.waves }}

{{ .name }}

{{ .what | safeHTML }}

Homes: {{ .homes }}

{{ end }}
-

The {{ len .Site.Params.neworgs }} new organisations. The first four have a landing page and documentation; the others are on GitHub only.

-
    - {{ range .Site.Params.neworgs }} -
  • {{ .name }} ({{ .state }}): {{ .what }}{{ with .site }} · landing{{ end }}{{ with .docs }} · docs{{ end }}
  • - {{ end }} -
+

The organisations of the plan are in the organisations with the others, each card with its status: {{ len (where $orgs "live" true) }} have a landing and documentation, and the {{ len (where $orgs "live" "!=" true) }} planned ones link to GitHub and the plan.

diff --git a/layouts/partials/org-logo.html b/layouts/partials/org-logo.html new file mode 100644 index 0000000..7417da7 --- /dev/null +++ b/layouts/partials/org-logo.html @@ -0,0 +1,11 @@ +{{- /* An organisation's logo for the cards and the stack diagram: its own + brand when published (`brand = true` in hugo.toml), otherwise the + go-scikit mark, faded, so that every card has the same shape and none + shows a broken image. Context: dict "org" (a params.orgs entry), "size" + and "radius" in pixels. */ -}} +{{- $o := .org -}} +{{- if $o.brand -}} +{{ $o.name }} logo +{{- else -}} +{{ $o.name }} (no logo yet) +{{- end -}}