generalized

skcriteria/plot/__init__.py

@@ -55,6 +55,10 @@
 
 import six
 
+import numpy as np
+
+from six.moves import zip
+
 from .. import norm
 from .radar import radar_plot
 from .multihist import multihist_plot
@@ -97,21 +101,38 @@ def __call__(self, plotname="scatter_matrix", **kwargs):
     def to_str(self):
         return "PlotProxy for {}".format(self._data)
 
-    def preprocess(self, data, mnorm, wnorm):
+    def preprocess(self, data, mnorm, wnorm, weighted):
+        # normalization
         nmtx = norm.norm(mnorm, data.mtx, criteria=data.criteria, axis=0)
         nweights = (
             norm.norm(wnorm, data.weights, criteria=data.criteria)
             if data.weights is not None else None)
-        return nmtx, data.criteria, nweights
 
-    def plot(self, func, mnorm="none", wnorm="none", **kwargs):
-        data = self._data
-        nmtx, criteria, nweights = self.preprocess(data, mnorm, wnorm)
+        # weight the data
+        if weighted and nweights is not None:
+            wmtx = np.multiply(nmtx, nweights)
+        else:
+            wmtx = nmtx
+
+        # labels for criteria
+        if nweights is not None:
+            clabels = [
+                "{} (w.{:.2f})".format(cn, cw)
+                for cn, cw in zip(data.cnames, nweights)]
+        else:
+            clabels = ["{}".format(cn) for cn in data.cnames]
+
+        return wmtx, data.criteria, nweights, data.anames, clabels
+
+    def plot(self, func, mnorm="none", wnorm="none", weighted=True, **kwargs):
+        mtx, criteria, weights, anames, cnames, = self.preprocess(
+            self._data, mnorm, wnorm, weighted)
+
         kwargs.update({
-            "mtx": nmtx, "criteria": criteria,
-            "weights": nweights,
-            "anames": kwargs.get("anames", data.anames),
-            "cnames": kwargs.get("cnames", data.cnames)})
+            "mtx": mtx, "criteria": criteria,
+            "weights": weights,
+            "anames": kwargs.pop("anames", anames),
+            "cnames": kwargs.pop("cnames", cnames)})
         return func(**kwargs)
 
     def radar(self, **kwargs):

skcriteria/plot/multihist.py

@@ -54,8 +54,7 @@
 
 def multihist_plot(
         mtx, criteria, weights, anames, cnames,
-        weighted=True, cmap=None, ax=None,
-        subplots_kwargs=None, hist_kwargs=None):
+        cmap=None, ax=None, subplots_kwargs=None, hist_kwargs=None):
 
     # create ax if necesary
     if ax is None:
@@ -66,31 +65,17 @@ def multihist_plot(
     cmap = cm.get_cmap(name=cmap)
     colors = cmap(np.linspace(0, 1, mtx.shape[1]))
 
-    # weight the data
-    if weighted and weights is not None:
-        wdata = np.multiply(mtx, weights)
-    else:
-        wdata = mtx
-
     # histogram
     hist_kwargs = hist_kwargs or {}
     hist_kwargs.setdefault("histtype", "stepfilled")
     hist_kwargs.setdefault("alpha", 0.8)
 
-    for arr, col in zip(wdata.T, colors):
+    for arr, col in zip(mtx.T, colors):
         ax.hist(arr, color=col, **hist_kwargs)
 
     ax.spines['right'].set_visible(False)
     ax.spines['top'].set_visible(False)
 
-    # labels for criteria
-    if weights is not None:
-        clabels = [
-            "{} (w.{:.2f})".format(cn, cw)
-            for cn, cw in zip(cnames, weights)]
-    else:
-        clabels = ["{}".format(cn) for cn in cnames]
-
-    ax.legend(clabels, loc="best")
+    ax.legend(cnames, loc="best")
 
     return ax

skcriteria/plot/radar.py

@@ -165,7 +165,7 @@ def unit_poly_verts(theta):
     return verts
 
 
-def radar_plot(mtx, criteria, weights, anames, cnames, weighted=True,
+def radar_plot(mtx, criteria, weights, anames, cnames,
                frame="polygon", cmap=None, ax=None, labelrow=5,
                subplots_kwargs=None):
 
@@ -183,35 +183,23 @@ def radar_plot(mtx, criteria, weights, anames, cnames, weighted=True,
     mincrits = np.squeeze(np.where(criteria == util.MIN))
     if np.any(mincrits):
         mincrits_inverted = 1.0 / mtx[:, mincrits]
-        pdata = mtx.astype(mincrits_inverted.dtype.type)
-        pdata[:, mincrits] = mincrits_inverted
+        pmtx = mtx.astype(mincrits_inverted.dtype.type)
+        pmtx[:, mincrits] = mincrits_inverted
     else:
-        pdata = mtx
-
-    # weight the data
-    if weighted and weights is not None:
-        wdata = np.multiply(pdata, weights)
-    else:
-        wdata = pdata
+        pmtx = mtx
 
     # colors
     cmap = cm.get_cmap(name=cmap)
     colors = cmap(np.linspace(0, 1, mtx.shape[0]))
 
     # Plot
     ax.set_rgrids([0.5])
-    for d, color in zip(norm.sum(wdata, axis=0), colors):
+    for d, color in zip(norm.sum(pmtx, axis=0), colors):
         ax.plot(theta, d, color=color)
         ax.fill(theta, d, facecolor=color, alpha=0.25)
 
     # labels for criteria
-    if weights is not None:
-        clabels = [
-            "{}\n(w.{:.2f})".format(cn, cw)
-            for cn, cw in zip(cnames, weights)]
-    else:
-        clabels = ["{}".format(cn) for cn in cnames]
-    ax.set_varlabels(clabels)
+    ax.set_varlabels(cnames)
 
     # legend
     colfactor = 1

skcriteria/plot/scmtx.py

@@ -51,7 +51,7 @@
 import matplotlib.pyplot as plt
 from matplotlib import cm
 
-from six.moves import range, zip
+from six.moves import range
 
 
 # =============================================================================
@@ -61,7 +61,7 @@
 def scatter_plot_matrix(
         data, labels, colors, axes=None,
         subplots_kwargs=None, scatter_kwargs=None, hist_kwargs=None):
-    """ Create a scatter plot matrix from the given data.
+    """Create a scatter plot matrix from the given data.
 
         Parameters
         ----------
@@ -153,25 +153,11 @@ def scatter_plot_matrix(
     return axes
 
 
-def scmtx_plot(mtx, criteria, weights, anames, cnames, weighted=True,
+def scmtx_plot(mtx, criteria, weights, anames, cnames,
                frame="polygon", cmap=None, ax=None, **kwargs):
 
     cmap = cm.get_cmap(name=cmap)
     colors = cmap(np.linspace(0, 1, mtx.shape[1] ** 2))
 
-    # weight the data
-    if weighted and weights is not None:
-        wdata = np.multiply(mtx, weights)
-    else:
-        wdata = mtx
-
-    # labels for criteria
-    if weights is not None:
-        clabels = [
-            "{} (w.{:.2f})".format(cn, cw)
-            for cn, cw in zip(cnames, weights)]
-    else:
-        clabels = ["{}".format(cn) for cn in cnames]
-
     return scatter_plot_matrix(
-        wdata.T, labels=clabels, colors=colors, axes=ax, **kwargs)
+        mtx.T, labels=cnames, colors=colors, axes=ax, **kwargs)