biplotpy 0.4.0__py3-none-any.whl

biplotpy/classical.py

@@ -0,0 +1,159 @@
+"""Classical biplots: GH (CMP), JK (RMP), SQRT and HJ-Biplot.
+
+Port of ``PCA.Biplot`` and ``HJ.Biplot`` (MultBiplotR, Vicente-Villardón).
+With the singular value decomposition X = U D V' of the transformed data:
+
+* alpha = 1  : JK / RMP-Biplot   rows A = U D,          columns B = V
+* alpha = 0  : GH / CMP-Biplot   rows A = U,            columns B = V D
+* alpha = 0.5: SQRT biplot       rows A = U D^{1/2},    columns B = V D^{1/2}
+* HJ-Biplot (Galindo, 1986): A = U D and B = V D, both with optimal quality.
+
+For alpha <= 1 the markers are rescaled by a common factor so that rows and
+columns have comparable spread (``scale_factor``), as in MultBiplotR.
+"""
+from __future__ import annotations
+
+import numpy as np
+import pandas as pd
+
+from .transforms import initial_transform, transform_name
+
+
+class ContinuousBiplot:
+    """Result of a classical biplot (fields mirror MultBiplotR's ``ContinuousBiplot``)."""
+
+    def __init__(self, **kw):
+        self.__dict__.update(kw)
+
+    # labelled views ---------------------------------------------------------
+    def _df(self, M, idx):
+        return pd.DataFrame(M, index=idx, columns=[f"Dim {d + 1}" for d in range(M.shape[1])])
+
+    @property
+    def rows(self):
+        return self._df(self.row_coordinates, self.row_names)
+
+    @property
+    def columns(self):
+        return self._df(self.col_coordinates, self.col_names)
+
+    def qualities(self, which="columns"):
+        """Quality of representation: cumulative relative contributions (%)."""
+        if which == "columns":
+            return self._df(np.cumsum(self.col_contributions, 1), self.col_names)
+        return self._df(np.cumsum(self.row_contributions, 1), self.row_names)
+
+    def summary(self) -> dict:
+        d = self.dim
+        eig = pd.DataFrame({"Eigenvalue": self.eigenvalues[:d], "Exp. Var": self.inertia[:d],
+                            "Cumulative": self.cum_inertia[:d]}, index=[f"Dim {i + 1}" for i in range(d)])
+        return {"eigen": eig,
+                "col_contributions": self._df(self.col_contributions, self.col_names),
+                "col_qualities": self.qualities("columns"),
+                "row_contributions": self._df(self.row_contributions, self.row_names),
+                "row_qualities": self.qualities("rows"),
+                "structure": self._df(self.structure, self.col_names),
+                "row_contrib_to_factor": self._df(self.row_contrib_to_factor, self.row_names),
+                "col_contrib_to_factor": self._df(self.col_contrib_to_factor, self.col_names)}
+
+    def plot(self, *args, **kwargs):
+        from .plot import plot_biplot
+        return plot_biplot(self, *args, **kwargs)
+
+    def __repr__(self):
+        s = self.summary()
+        return (f"{self.title} ({self.normalization}); transformation: {self.transformation}\n"
+                f"{s['eigen'].round(3).to_string()}\n\nQuality of representation of the columns (%):\n"
+                f"{s['col_qualities'].round(2).to_string()}")
+
+
+_NORMALIZATION = {2: "Principal normalization (HJ-Biplot)", 1: "Row principal normalization (JK / RMP-Biplot)",
+                  0: "Column principal normalization (GH / CMP-Biplot)", 0.5: "Symmetrical normalization (SQRT-Biplot)"}
+
+
+def pca_biplot(X, alpha=1.0, dim=2, scaling=5, sup_rows=None, sup_cols=None, grouping=None,
+               row_names=None, col_names=None) -> ContinuousBiplot:
+    """Classical PCA biplot with normalization ``alpha`` (``alpha > 1`` gives the HJ-Biplot).
+
+    ``scaling`` is the initial transformation (name or MultBiplotR number; 5 =
+    standardize columns, 4 = column centering).
+    """
+    if isinstance(X, pd.DataFrame):
+        row_names = row_names or [str(i) for i in X.index]
+        col_names = col_names or [str(c) for c in X.columns]
+        X = X.to_numpy(float)
+    X0 = np.array(X, float)
+    n, p = X0.shape
+    row_names = list(row_names) if row_names is not None else [f"I{i + 1}" for i in range(n)]
+    col_names = list(col_names) if col_names is not None else [f"V{j + 1}" for j in range(p)]
+    tr = initial_transform(X0, transform_name(scaling), sup_rows, sup_cols, grouping)
+    Xs = tr["X"]
+    U, d, Vt = np.linalg.svd(Xs, full_matrices=False)
+    V = Vt.T
+    ev = d ** 2
+    inertia = ev / ev.sum() * 100
+    a = U[:, :dim] * d[:dim]
+    b = V[:, :dim] * d[:dim]
+    structure = np.corrcoef(np.column_stack([Xs, a]), rowvar=False)[:p, p:]
+    sf = (Xs ** 2).sum(1)
+    rc = a ** 2 / sf[:, None] * 100
+    scc = (Xs ** 2).sum(0)
+    cc = b ** 2 / scc[:, None] * 100
+    # contributions of each element to the factor (absolute contributions, %)
+    rcf = a ** 2 / (a ** 2).sum(0) * 100
+    ccf = b ** 2 / (b ** 2).sum(0) * 100
+    sr, sc = tr["sup_rows"], tr["sup_cols"]
+    a_s = b_s = None
+    if sr is not None:
+        a_s = sr @ V[:, :dim]
+        rcs = a_s ** 2 / (sr ** 2).sum(1)[:, None] * 100
+    if sc is not None:
+        b_s = sc.T @ U[:, :dim]
+        ccs = b_s ** 2 / (sc ** 2).sum(0)[:, None] * 100
+    if alpha <= 1:
+        a = a * (1 / d[:dim]) ** (1 - alpha)
+        b = b * (1 / d[:dim]) ** alpha
+        if a_s is not None:
+            a_s = a_s * (1 / d[:dim]) ** (1 - alpha)
+        if b_s is not None:
+            b_s = b_s * (1 / d[:dim]) ** alpha
+        sca = (a ** 2).sum() / n
+        scb = (b ** 2).sum() / p
+        scf = np.sqrt(np.sqrt(scb / sca))
+        a, b = a * scf, b / scf
+        a_s = None if a_s is None else a_s * scf
+        b_s = None if b_s is None else b_s / scf
+    else:
+        scf = 1.0
+    norm = _NORMALIZATION.get(alpha if alpha <= 1 else 2, f"Custom normalization (alpha = {alpha})")
+    return ContinuousBiplot(
+        title="HJ Biplot" if alpha > 1 else "PCA Biplot", type="HJ" if alpha > 1 else "PCA", alpha=alpha,
+        normalization=norm, dim=dim, transformation=tr["transform"], non_scaled_data=X0, scaled_data=Xs,
+        means=X0.mean(0), medians=np.median(X0, 0), deviations=tr["col_sds"], minima=X0.min(0), maxima=X0.max(0),
+        p25=np.quantile(X0, 0.25, axis=0), p75=np.quantile(X0, 0.75, axis=0), gmean=X0.mean(),
+        eigenvalues=ev, inertia=inertia, cum_inertia=np.cumsum(inertia), V=V[:, :dim], U=U[:, :dim], d=d,
+        structure=structure, row_coordinates=a, col_coordinates=b, row_contributions=rc, col_contributions=cc,
+        row_contrib_to_factor=rcf, col_contrib_to_factor=ccf, scale_factor=scf, row_names=row_names,
+        col_names=col_names, n=n, p=p,
+        sup_row_coordinates=a_s, sup_col_coordinates=b_s,
+        sup_row_contributions=None if sr is None else rcs, sup_col_contributions=None if sc is None else ccs)
+
+
+def hj_biplot(X, dim=3, scaling=5, **kw) -> ContinuousBiplot:
+    """HJ-Biplot (Galindo, 1986): rows and columns with optimal quality of representation."""
+    return pca_biplot(X, alpha=2, dim=dim, scaling=scaling, **kw)
+
+
+def gh_biplot(X, dim=2, scaling=5, **kw) -> ContinuousBiplot:
+    """GH-Biplot (column metric preserving, alpha = 0)."""
+    return pca_biplot(X, alpha=0, dim=dim, scaling=scaling, **kw)
+
+
+def jk_biplot(X, dim=2, scaling=5, **kw) -> ContinuousBiplot:
+    """JK-Biplot (row metric preserving, alpha = 1)."""
+    return pca_biplot(X, alpha=1, dim=dim, scaling=scaling, **kw)
+
+
+def sqrt_biplot(X, dim=2, scaling=5, **kw) -> ContinuousBiplot:
+    """Symmetrical (SQRT) biplot, alpha = 0.5."""
+    return pca_biplot(X, alpha=0.5, dim=dim, scaling=scaling, **kw)

biplotpy/correspondence.py

@@ -0,0 +1,231 @@
+"""Correspondence-type biplots: simple and multiple correspondence analysis,
+canonical correspondence analysis (CCA) and redundancy analysis (RDA).
+
+* ``ca``  — port of ``MultBiplotR::CA`` (alpha = 1 rows principal / columns
+  standard, 0 the reverse, 0.5 symmetric biplot, 2 = both principal, the
+  usual "symmetric map" of ``ca``/``FactoMineR``); also returns the absolute
+  contributions (to the axis) of ``ca::ca``/``FactoMineR::CA``.
+* ``mca`` — multiple correspondence analysis as the CA of the indicator matrix
+  (``FactoMineR::MCA`` conventions), with the Benzécri and Greenacre adjusted
+  inertias.
+* ``cca`` — port of ``MultBiplotR::CCA`` (ter Braak, 1986), eigenvalues equal
+  to ``vegan::cca``.
+* ``rda`` — redundancy analysis with the scores and scalings of ``vegan::rda``.
+"""
+from __future__ import annotations
+
+import numpy as np
+import pandas as pd
+
+from .transforms import initial_transform, transform_name
+
+
+class OrdinationBiplot:
+    """Result of a CA/MCA/CCA/RDA/PCoA/MDS-type biplot."""
+
+    def __init__(self, **kw):
+        self.__dict__.update(kw)
+
+    def __repr__(self):
+        ev = np.asarray(self.eigenvalues)[:5]
+        return f"{self.title}: eigenvalues {np.round(ev, 4)}; inertia (%) {np.round(np.asarray(self.inertia)[:5], 2)}"
+
+    def plot(self, *args, **kwargs):
+        from .plot import plot_ordination
+        return plot_ordination(self, *args, **kwargs)
+
+
+def _names(x, n, p, rp="I", cp="V"):
+    if isinstance(x, pd.DataFrame):
+        return [str(i) for i in x.index], [str(c) for c in x.columns]
+    return [f"{rp}{i + 1}" for i in range(n)], [f"{cp}{j + 1}" for j in range(p)]
+
+
+# --------------------------------------------------------------------------- CA
+def ca(N, dim=2, alpha=2.0, sup_rows=None, sup_cols=None) -> OrdinationBiplot:
+    """Simple correspondence analysis of a contingency table ``N``.
+
+    ``alpha``: 2 = symmetric map (rows and columns in principal coordinates);
+    1 = row principal (RMP); 0 = column principal (CMP); 0.5 = symmetric
+    biplot (MultBiplotR's convention for alpha <= 1)."""
+    rn, cn = _names(N, *np.shape(N))
+    X = np.asarray(N, float)
+    n, p = X.shape
+    tot = X.sum()
+    P = X / tot
+    r, c = P.sum(1), P.sum(0)
+    S = (P - np.outer(r, c)) / np.sqrt(np.outer(r, c))
+    U, d, Vt = np.linalg.svd(S, full_matrices=False)
+    k = min(n, p) - 1
+    U, d, V = U[:, :k], d[:k], Vt.T[:, :k]
+    Fs, Gs = U / np.sqrt(r)[:, None], V / np.sqrt(c)[:, None]        # standard coordinates
+    F, G = Fs * d, Gs * d                                              # principal coordinates
+    ev = d ** 2
+    inertia = ev / ev.sum() * 100
+    qual_r = F ** 2 / (F ** 2).sum(1, keepdims=True) * 100
+    qual_c = G ** 2 / (G ** 2).sum(1, keepdims=True) * 100
+    ctr_r = r[:, None] * F ** 2 / ev * 100
+    ctr_c = c[:, None] * G ** 2 / ev * 100
+    if alpha > 1:
+        A, B = F, G
+    else:
+        A, B = Fs * d ** alpha, Gs * d ** (1 - alpha)
+    res = OrdinationBiplot(title="Correspondence Analysis", type="CA", alpha=alpha, dim=dim, eigenvalues=ev,
+                           inertia=inertia, cum_inertia=np.cumsum(inertia), total_inertia=ev.sum(),
+                           chi2=ev.sum() * tot, row_coordinates=A[:, :dim], col_coordinates=B[:, :dim],
+                           row_principal=F, col_principal=G, row_standard=Fs, col_standard=Gs,
+                           row_masses=r, col_masses=c, row_qualities=qual_r[:, :dim], col_qualities=qual_c[:, :dim],
+                           row_contributions=ctr_r[:, :dim], col_contributions=ctr_c[:, :dim],
+                           row_names=rn, col_names=cn, data=X)
+    if sup_rows is not None:
+        R = np.asarray(sup_rows, float); R = R / R.sum(1, keepdims=True)
+        res.sup_row_coordinates = (R @ Gs)[:, :dim]
+    if sup_cols is not None:
+        C = np.asarray(sup_cols, float); C = C / C.sum(0, keepdims=True)
+        res.sup_col_coordinates = (C.T @ Fs)[:, :dim]
+    return res
+
+
+# --------------------------------------------------------------------------- MCA
+def indicator_matrix(X):
+    """Complete disjunctive table of a data frame of categorical variables."""
+    df = pd.DataFrame(X).astype(str)
+    cols, levs, var_of = [], [], []
+    for v in df.columns:
+        for lev in sorted(df[v].unique()):
+            cols.append((df[v] == lev).astype(float).to_numpy())
+            levs.append(lev); var_of.append(v)
+    # category labels as in FactoMineR: the level itself when unique, else variable_level
+    names = levs if len(set(levs)) == len(levs) else [f"{v}_{l}" for v, l in zip(var_of, levs)]
+    return np.column_stack(cols), names, var_of
+
+
+def mca(X, dim=2, sup_rows=None) -> OrdinationBiplot:
+    """Multiple correspondence analysis (CA of the indicator matrix).
+
+    Coordinates follow ``FactoMineR::MCA``: individuals and categories in
+    principal coordinates; eigenvalues of the indicator matrix; also the
+    Benzécri-corrected and Greenacre-adjusted percentages of inertia."""
+    df = pd.DataFrame(X)
+    rn = [str(i) for i in df.index]
+    Z, names, var_of = indicator_matrix(df)
+    Q = df.shape[1]
+    res = ca(pd.DataFrame(Z, index=rn, columns=names), dim=max(dim, 2), alpha=2)
+    ev = res.eigenvalues
+    keep = ev > 1 / Q
+    benz = ((Q / (Q - 1)) * (ev[keep] - 1 / Q)) ** 2
+    J = Z.shape[1]
+    green_total = Q / (Q - 1) * (np.sum(ev ** 2) - (J - Q) / Q ** 2)
+    res.title, res.type = "Multiple Correspondence Analysis", "MCA"
+    res.row_coordinates, res.col_coordinates = res.row_principal[:, :dim], res.col_principal[:, :dim]
+    res.benzecri = benz / benz.sum() * 100
+    res.greenacre = benz / green_total * 100
+    res.variables = var_of
+    res.eta2 = np.array([[np.sum(res.col_masses[np.array(var_of) == v] * res.col_principal[np.array(var_of) == v, k] ** 2)
+                          * Q for k in range(dim)] for v in df.columns])
+    res.var_names = list(df.columns)
+    return res
+
+
+# --------------------------------------------------------------------------- CCA
+def _msqrt(S, inv=False):
+    w, U = np.linalg.eigh(S)
+    w = np.maximum(w, 0)
+    return U @ np.diag(1 / np.sqrt(w) if inv else np.sqrt(w)) @ U.T
+
+
+def cca(Y, Z, alpha=1.0, dim=2) -> OrdinationBiplot:
+    """Canonical correspondence analysis of an abundance table ``Y`` (sites x
+    species) constrained by environmental variables ``Z`` (sites x variables).
+
+    Port of ``MultBiplotR::CCA``: species scores, site scores (weighted
+    averages, WA) and linear-combination site scores (LC), environmental
+    variable scores (biplot arrows), qualities and inertia decompositions.
+    ``alpha`` = 1 gives species in principal coordinates."""
+    sn, spn = _names(Y, *np.shape(Y), "Si", "Sp")
+    en = list(Z.columns) if isinstance(Z, pd.DataFrame) else [f"Ev{k + 1}" for k in range(np.shape(Z)[1])]
+    P = np.asarray(Y, float); Zm = np.asarray(Z, float)
+    P = P / P.sum()
+    r, c = P.sum(1), P.sum(0)
+    PC = (P - np.outer(r, c)) / np.sqrt(np.outer(r, c))
+    valsin = np.linalg.svd(PC, compute_uv=False)
+    in_total = np.sum(PC ** 2)
+    iner_rows, iner_cols = (PC ** 2).sum(1), (PC ** 2).sum(0)
+    W1 = (P.T @ Zm) / c[:, None]
+    mean_z = r @ Zm
+    Zc = Zm - mean_z
+    s11 = r @ Zc ** 2
+    Zs = Zc / np.sqrt(s11)
+    W = (P.T @ Zs) / c[:, None]
+    S22 = Zs.T @ (r[:, None] * Zs)
+    sq, isq = _msqrt(S22), _msqrt(S22, inv=True)
+    WW = np.sqrt(c)[:, None] * W @ isq
+    U, d, Vt = np.linalg.svd(WW, full_matrices=False)
+    V = Vt.T
+    x = (P / r[:, None]) @ (U / np.sqrt(c)[:, None])
+    x2 = Zs @ isq @ V
+    A = (U / np.sqrt(c)[:, None]) * d ** alpha
+    B = sq @ V * d ** (1 - alpha)
+    Afull = (U / np.sqrt(c)[:, None]) * d
+    Bfull = sq @ V * d
+    sp_q = Afull ** 2 / (Afull ** 2).sum(1, keepdims=True) * 100
+    ev_q = Bfull ** 2 / (Bfull ** 2).sum(1, keepdims=True) * 100
+    site_q = (np.sqrt(r)[:, None] * x) ** 2 / iner_rows[:, None] * 100
+    sp_q_ca = (U * d) ** 2 / iner_cols[:, None] * 100
+    interset = np.corrcoef(np.column_stack([x, Zm]), rowvar=False)[:x.shape[1], x.shape[1]:]
+    ev = d ** 2
+    return OrdinationBiplot(title="Canonical Correspondence Analysis", type="CCA", alpha=alpha, dim=dim,
+                            eigenvalues=ev, inertia=ev / ev.sum() * 100, exp_species_constrained=ev / in_total * 100,
+                            exp_species_ca=valsin ** 2 / in_total * 100, total_inertia=in_total,
+                            row_coordinates=(x * d ** (-alpha))[:, :dim], lc_site_scores=(x2 * d ** (-alpha))[:, :dim],
+                            col_coordinates=A[:, :dim], env_coordinates=B[:, :dim], weighted_averages=W1,
+                            species_quality=sp_q[:, :dim], species_quality_ca=sp_q_ca[:, :dim],
+                            env_quality=ev_q[:, :dim], site_quality=site_q[:, :dim], interset=interset[:, :dim],
+                            row_names=sn, col_names=spn, env_names=en, means_z=mean_z, sd_z=np.sqrt(s11))
+
+
+# --------------------------------------------------------------------------- RDA
+def rda(Y, X, scale=False, scaling=2, dim=2) -> OrdinationBiplot:
+    """Redundancy analysis (constrained PCA) of ``Y`` on ``X``.
+
+    Scores and scalings follow ``vegan::rda``/``scores.rda`` (scaling 1 =
+    sites, 2 = species, 3 = symmetric; constant sqrt(sqrt((n-1) * total
+    variance))): site scores (WA), constrained site scores (LC), species scores
+    and biplot scores of the explanatory variables; unconstrained (residual)
+    axes are returned as ``ca_eigenvalues``."""
+    sn, spn = _names(Y, *np.shape(Y), "S", "Sp")
+    xn = list(X.columns) if isinstance(X, pd.DataFrame) else [f"X{k + 1}" for k in range(np.shape(X)[1])]
+    Ym = np.asarray(Y, float); Xm = np.asarray(X, float)
+    n = Ym.shape[0]
+    Yc = Ym - Ym.mean(0)
+    if scale:
+        Yc = Yc / Ym.std(0, ddof=1)
+    Ys = Yc / np.sqrt(n - 1)
+    tot = np.sum(Ys ** 2)
+    Xc = Xm - Xm.mean(0)
+    Qx, _ = np.linalg.qr(Xc)
+    rank = np.linalg.matrix_rank(Xc)
+    Qx = Qx[:, :rank]
+    Fit = Qx @ (Qx.T @ Ys)
+    U, d, Vt = np.linalg.svd(Fit, full_matrices=False)
+    k = int(np.sum(d > 1e-10 * d[0]))
+    U, d, V = U[:, :k], d[:k], Vt.T[:, :k]
+    ev = d ** 2
+    wa = Ys @ V / d
+    biplot = np.corrcoef(np.column_stack([Xm, U]), rowvar=False)[:Xm.shape[1], Xm.shape[1]:]
+    Res = Ys - Fit
+    res_ev = np.linalg.svd(Res, compute_uv=False) ** 2
+    res_ev = res_ev[res_ev > 1e-10 * max(res_ev.max(), 1e-300)]
+    const = np.sqrt(np.sqrt((n - 1) * tot))
+    eig_all = np.concatenate([ev, res_ev])
+    slam = np.sqrt(ev / tot)
+    sp_s, site_s = {1: (1, slam), 2: (slam, 1), 3: (np.sqrt(slam), np.sqrt(slam))}[scaling]
+    species = const * V * sp_s
+    sites = const * wa * site_s
+    lc = const * U * site_s
+    bp = biplot * site_s
+    return OrdinationBiplot(title="Redundancy Analysis", type="RDA", dim=dim, scaling=scaling, eigenvalues=ev,
+                            inertia=ev / tot * 100, ca_eigenvalues=res_ev, total_inertia=tot,
+                            constrained_proportion=ev.sum() / tot, row_coordinates=sites[:, :dim],
+                            lc_site_scores=lc[:, :dim], col_coordinates=species[:, :dim], env_coordinates=bp[:, :dim],
+                            row_names=sn, col_names=spn, env_names=xn, const=const, all_eigenvalues=eig_all)

biplotpy/datasets/__init__.py

@@ -0,0 +1,19 @@
+"""Classic datasets used in the examples."""
+from importlib import resources
+
+import pandas as pd
+
+
+def _read(name):
+    with resources.files(__package__).joinpath(name).open("r") as f:
+        return pd.read_csv(f)
+
+
+def load_iris() -> pd.DataFrame:
+    """Fisher's iris data (150 x 5)."""
+    return _read("iris.csv")
+
+
+def load_mtcars() -> pd.DataFrame:
+    """Motor Trend cars (32 x 12, with a ``name`` column)."""
+    return _read("mtcars.csv").set_index("name")

biplotpy/datasets/iris.csv

@@ -0,0 +1,151 @@
+"Sepal.Length","Sepal.Width","Petal.Length","Petal.Width","Species"
+5.1,3.5,1.4,0.2,"setosa"
+4.9,3,1.4,0.2,"setosa"
+4.7,3.2,1.3,0.2,"setosa"
+4.6,3.1,1.5,0.2,"setosa"
+5,3.6,1.4,0.2,"setosa"
+5.4,3.9,1.7,0.4,"setosa"
+4.6,3.4,1.4,0.3,"setosa"
+5,3.4,1.5,0.2,"setosa"
+4.4,2.9,1.4,0.2,"setosa"
+4.9,3.1,1.5,0.1,"setosa"
+5.4,3.7,1.5,0.2,"setosa"
+4.8,3.4,1.6,0.2,"setosa"
+4.8,3,1.4,0.1,"setosa"
+4.3,3,1.1,0.1,"setosa"
+5.8,4,1.2,0.2,"setosa"
+5.7,4.4,1.5,0.4,"setosa"
+5.4,3.9,1.3,0.4,"setosa"
+5.1,3.5,1.4,0.3,"setosa"
+5.7,3.8,1.7,0.3,"setosa"
+5.1,3.8,1.5,0.3,"setosa"
+5.4,3.4,1.7,0.2,"setosa"
+5.1,3.7,1.5,0.4,"setosa"
+4.6,3.6,1,0.2,"setosa"
+5.1,3.3,1.7,0.5,"setosa"
+4.8,3.4,1.9,0.2,"setosa"
+5,3,1.6,0.2,"setosa"
+5,3.4,1.6,0.4,"setosa"
+5.2,3.5,1.5,0.2,"setosa"
+5.2,3.4,1.4,0.2,"setosa"
+4.7,3.2,1.6,0.2,"setosa"
+4.8,3.1,1.6,0.2,"setosa"
+5.4,3.4,1.5,0.4,"setosa"
+5.2,4.1,1.5,0.1,"setosa"
+5.5,4.2,1.4,0.2,"setosa"
+4.9,3.1,1.5,0.2,"setosa"
+5,3.2,1.2,0.2,"setosa"
+5.5,3.5,1.3,0.2,"setosa"
+4.9,3.6,1.4,0.1,"setosa"
+4.4,3,1.3,0.2,"setosa"
+5.1,3.4,1.5,0.2,"setosa"
+5,3.5,1.3,0.3,"setosa"
+4.5,2.3,1.3,0.3,"setosa"
+4.4,3.2,1.3,0.2,"setosa"
+5,3.5,1.6,0.6,"setosa"
+5.1,3.8,1.9,0.4,"setosa"
+4.8,3,1.4,0.3,"setosa"
+5.1,3.8,1.6,0.2,"setosa"
+4.6,3.2,1.4,0.2,"setosa"
+5.3,3.7,1.5,0.2,"setosa"
+5,3.3,1.4,0.2,"setosa"
+7,3.2,4.7,1.4,"versicolor"
+6.4,3.2,4.5,1.5,"versicolor"
+6.9,3.1,4.9,1.5,"versicolor"
+5.5,2.3,4,1.3,"versicolor"
+6.5,2.8,4.6,1.5,"versicolor"
+5.7,2.8,4.5,1.3,"versicolor"
+6.3,3.3,4.7,1.6,"versicolor"
+4.9,2.4,3.3,1,"versicolor"
+6.6,2.9,4.6,1.3,"versicolor"
+5.2,2.7,3.9,1.4,"versicolor"
+5,2,3.5,1,"versicolor"
+5.9,3,4.2,1.5,"versicolor"
+6,2.2,4,1,"versicolor"
+6.1,2.9,4.7,1.4,"versicolor"
+5.6,2.9,3.6,1.3,"versicolor"
+6.7,3.1,4.4,1.4,"versicolor"
+5.6,3,4.5,1.5,"versicolor"
+5.8,2.7,4.1,1,"versicolor"
+6.2,2.2,4.5,1.5,"versicolor"
+5.6,2.5,3.9,1.1,"versicolor"
+5.9,3.2,4.8,1.8,"versicolor"
+6.1,2.8,4,1.3,"versicolor"
+6.3,2.5,4.9,1.5,"versicolor"
+6.1,2.8,4.7,1.2,"versicolor"
+6.4,2.9,4.3,1.3,"versicolor"
+6.6,3,4.4,1.4,"versicolor"
+6.8,2.8,4.8,1.4,"versicolor"
+6.7,3,5,1.7,"versicolor"
+6,2.9,4.5,1.5,"versicolor"
+5.7,2.6,3.5,1,"versicolor"
+5.5,2.4,3.8,1.1,"versicolor"
+5.5,2.4,3.7,1,"versicolor"
+5.8,2.7,3.9,1.2,"versicolor"
+6,2.7,5.1,1.6,"versicolor"
+5.4,3,4.5,1.5,"versicolor"
+6,3.4,4.5,1.6,"versicolor"
+6.7,3.1,4.7,1.5,"versicolor"
+6.3,2.3,4.4,1.3,"versicolor"
+5.6,3,4.1,1.3,"versicolor"
+5.5,2.5,4,1.3,"versicolor"
+5.5,2.6,4.4,1.2,"versicolor"
+6.1,3,4.6,1.4,"versicolor"
+5.8,2.6,4,1.2,"versicolor"
+5,2.3,3.3,1,"versicolor"
+5.6,2.7,4.2,1.3,"versicolor"
+5.7,3,4.2,1.2,"versicolor"
+5.7,2.9,4.2,1.3,"versicolor"
+6.2,2.9,4.3,1.3,"versicolor"
+5.1,2.5,3,1.1,"versicolor"
+5.7,2.8,4.1,1.3,"versicolor"
+6.3,3.3,6,2.5,"virginica"
+5.8,2.7,5.1,1.9,"virginica"
+7.1,3,5.9,2.1,"virginica"
+6.3,2.9,5.6,1.8,"virginica"
+6.5,3,5.8,2.2,"virginica"
+7.6,3,6.6,2.1,"virginica"
+4.9,2.5,4.5,1.7,"virginica"
+7.3,2.9,6.3,1.8,"virginica"
+6.7,2.5,5.8,1.8,"virginica"
+7.2,3.6,6.1,2.5,"virginica"
+6.5,3.2,5.1,2,"virginica"
+6.4,2.7,5.3,1.9,"virginica"
+6.8,3,5.5,2.1,"virginica"
+5.7,2.5,5,2,"virginica"
+5.8,2.8,5.1,2.4,"virginica"
+6.4,3.2,5.3,2.3,"virginica"
+6.5,3,5.5,1.8,"virginica"
+7.7,3.8,6.7,2.2,"virginica"
+7.7,2.6,6.9,2.3,"virginica"
+6,2.2,5,1.5,"virginica"
+6.9,3.2,5.7,2.3,"virginica"
+5.6,2.8,4.9,2,"virginica"
+7.7,2.8,6.7,2,"virginica"
+6.3,2.7,4.9,1.8,"virginica"
+6.7,3.3,5.7,2.1,"virginica"
+7.2,3.2,6,1.8,"virginica"
+6.2,2.8,4.8,1.8,"virginica"
+6.1,3,4.9,1.8,"virginica"
+6.4,2.8,5.6,2.1,"virginica"
+7.2,3,5.8,1.6,"virginica"
+7.4,2.8,6.1,1.9,"virginica"
+7.9,3.8,6.4,2,"virginica"
+6.4,2.8,5.6,2.2,"virginica"
+6.3,2.8,5.1,1.5,"virginica"
+6.1,2.6,5.6,1.4,"virginica"
+7.7,3,6.1,2.3,"virginica"
+6.3,3.4,5.6,2.4,"virginica"
+6.4,3.1,5.5,1.8,"virginica"
+6,3,4.8,1.8,"virginica"
+6.9,3.1,5.4,2.1,"virginica"
+6.7,3.1,5.6,2.4,"virginica"
+6.9,3.1,5.1,2.3,"virginica"
+5.8,2.7,5.1,1.9,"virginica"
+6.8,3.2,5.9,2.3,"virginica"
+6.7,3.3,5.7,2.5,"virginica"
+6.7,3,5.2,2.3,"virginica"
+6.3,2.5,5,1.9,"virginica"
+6.5,3,5.2,2,"virginica"
+6.2,3.4,5.4,2.3,"virginica"
+5.9,3,5.1,1.8,"virginica"

biplotpy/datasets/mtcars.csv

@@ -0,0 +1,33 @@
+"mpg","cyl","disp","hp","drat","wt","qsec","vs","am","gear","carb","name"
+21,6,160,110,3.9,2.62,16.46,0,1,4,4,"Mazda RX4"
+21,6,160,110,3.9,2.875,17.02,0,1,4,4,"Mazda RX4 Wag"
+22.8,4,108,93,3.85,2.32,18.61,1,1,4,1,"Datsun 710"
+21.4,6,258,110,3.08,3.215,19.44,1,0,3,1,"Hornet 4 Drive"
+18.7,8,360,175,3.15,3.44,17.02,0,0,3,2,"Hornet Sportabout"
+18.1,6,225,105,2.76,3.46,20.22,1,0,3,1,"Valiant"
+14.3,8,360,245,3.21,3.57,15.84,0,0,3,4,"Duster 360"
+24.4,4,146.7,62,3.69,3.19,20,1,0,4,2,"Merc 240D"
+22.8,4,140.8,95,3.92,3.15,22.9,1,0,4,2,"Merc 230"
+19.2,6,167.6,123,3.92,3.44,18.3,1,0,4,4,"Merc 280"
+17.8,6,167.6,123,3.92,3.44,18.9,1,0,4,4,"Merc 280C"
+16.4,8,275.8,180,3.07,4.07,17.4,0,0,3,3,"Merc 450SE"
+17.3,8,275.8,180,3.07,3.73,17.6,0,0,3,3,"Merc 450SL"
+15.2,8,275.8,180,3.07,3.78,18,0,0,3,3,"Merc 450SLC"
+10.4,8,472,205,2.93,5.25,17.98,0,0,3,4,"Cadillac Fleetwood"
+10.4,8,460,215,3,5.424,17.82,0,0,3,4,"Lincoln Continental"
+14.7,8,440,230,3.23,5.345,17.42,0,0,3,4,"Chrysler Imperial"
+32.4,4,78.7,66,4.08,2.2,19.47,1,1,4,1,"Fiat 128"
+30.4,4,75.7,52,4.93,1.615,18.52,1,1,4,2,"Honda Civic"
+33.9,4,71.1,65,4.22,1.835,19.9,1,1,4,1,"Toyota Corolla"
+21.5,4,120.1,97,3.7,2.465,20.01,1,0,3,1,"Toyota Corona"
+15.5,8,318,150,2.76,3.52,16.87,0,0,3,2,"Dodge Challenger"
+15.2,8,304,150,3.15,3.435,17.3,0,0,3,2,"AMC Javelin"
+13.3,8,350,245,3.73,3.84,15.41,0,0,3,4,"Camaro Z28"
+19.2,8,400,175,3.08,3.845,17.05,0,0,3,2,"Pontiac Firebird"
+27.3,4,79,66,4.08,1.935,18.9,1,1,4,1,"Fiat X1-9"
+26,4,120.3,91,4.43,2.14,16.7,0,1,5,2,"Porsche 914-2"
+30.4,4,95.1,113,3.77,1.513,16.9,1,1,5,2,"Lotus Europa"
+15.8,8,351,264,4.22,3.17,14.5,0,1,5,4,"Ford Pantera L"
+19.7,6,145,175,3.62,2.77,15.5,0,1,5,6,"Ferrari Dino"
+15,8,301,335,3.54,3.57,14.6,0,1,5,8,"Maserati Bora"
+21.4,4,121,109,4.11,2.78,18.6,1,1,4,2,"Volvo 142E"