Add new (list) Statistics node (WIP)

docs/nodes/list_main/statistics.rst

@@ -0,0 +1,66 @@
+List Statistics
+===============
+
+Functionality
+-------------
+
+List Statistics node computes various statistical quantities for the values in a list.
+
+Inputs
+------
+
+The **Data** input is expected to be a list of integers / floats or list of lists of integers / floats.
+All inputs are vectorized.
+
+Parameters
+----------
+
+The **Function** parameter allows to select the statistical function to compute the corresponding statistical quantity for the input values.
+
++----------------+---------------------+---------+------------------------------------------+
+| Param          | Type                | Default | Description                              |
++================+=====================+=========+==========================================+
+| **Function**   | Enum                | Average | The statistical function applied to      |
+|                |  All Statistics     |         | the input values.                        |
+|                |  Sum                |         |                                          |
+|                |  Sum Of Squares     |         | For "All Statistics" selection the node  |
+|                |  Product            |         | computes and outputs the statistical     |
+|                |  Average            |         | quantities for all the statistical       |
+|                |  Geometric Mean     |         | functions along with their corresponding |
+|                |  Harmonic Mean      |         | names.                                   |
+|                |  Standard Deviation |         |                                          |
+|                |  Root Mean Square   |         |                                          |
+|                |  Skewness           |         |                                          |
+|                |  Kurtosis           |         |                                          |
+|                |  Minimum            |         |                                          |
+|                |  Maximum            |         |                                          |
+|                |  Median             |         |                                          |
+|                |  Percentile         |         |                                          |
+|                |  Histogram          |         |                                          |
++----------------+---------------------+---------+------------------------------------------+
+| **Percentage** | Float               | 0.75    | The percentage value for the             |
+|                |                     |         | percentile function. [1]                 |
++----------------+---------------------+---------+------------------------------------------+
+| **Normalize**  | Boolean             | False   | Flag to normalize the histogram bins     |
+|                |                     |         | to the given normalize size. [2]         |
++----------------+---------------------+---------+------------------------------------------+
+| **Bins**       | Int                 | 10      | The number of bins in the histogram. [2] |
++----------------+---------------------+---------+------------------------------------------+
+| **Size**       | Float               | 10.00   | The normalized size of the histogram.[2] |
++----------------+---------------------+---------+------------------------------------------+
+
+Notes:
+[1] : The **Percentage** input socket is available only for the **Percentile** function.
+[2] : The **Normalize** setting and the **Bins** and **Size** input sockets are available only for the **Histogram** function.
+
+Outputs
+-------
+**Name(s)**
+The name(s) of the statistical value(s) computed corresponding to the selected statistical function.
+
+**Value(s)**
+The statistical quantity of the input values corresponding to the selected function. For a vectorized input the output values are a series of quantities corresponding to the selected function.
+
+When "All Statistics" is selected the **Names** and **Values** outputs will list the names and the corresponding values for all the statistical functions.
+
+

index.md

@@ -146,6 +146,7 @@
     ListMatchNode
     ListFuncNode
     SvListDecomposeNode
+    SvListStatisticsNode
 
 ## List Struct
     ShiftNodeMK2

nodes/list_main/statistics.py

@@ -0,0 +1,188 @@
+# ##### BEGIN GPL LICENSE BLOCK #####
+#
+#  This program is free software; you can redistribute it and/or
+#  modify it under the terms of the GNU General Public License
+#  as published by the Free Software Foundation; either version 2
+#  of the License, or (at your option) any later version.
+#
+#  This program is distributed in the hope that it will be useful,
+#  but WITHOUT ANY WARRANTY; without even the implied warranty of
+#  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+#  GNU General Public License for more details.
+#
+#  You should have received a copy of the GNU General Public License
+#  along with this program; if not, write to the Free Software Foundation,
+#  Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
+#
+# ##### END GPL LICENSE BLOCK #####
+
+import bpy
+from bpy.props import EnumProperty, IntProperty, FloatProperty, BoolProperty
+
+from sverchok.node_tree import SverchCustomTreeNode
+from sverchok.data_structure import updateNode, match_long_repeat
+from sverchok.utils.modules.statistics_functions import *
+
+functions = {
+    "ALL STATISTICS":     (0, 0),
+    "SUM":                (10, get_sum),
+    "SUM OF SQUARES":     (11, get_sum_of_squares),
+    "SUM OF INVERSIONS":  (12, get_sum_of_inversions),
+    "PRODUCT":            (13, get_product),
+    "AVERAGE":            (14, get_average),
+    "GEOMETRIC MEAN":     (15, get_geometric_mean),
+    "HARMONIC MEAN":      (16, get_harmonic_mean),
+    "STANDARD DEVIATION": (17, get_standard_deviation),
+    "ROOT MEAN SQUARE":   (18, get_root_mean_square),
+    "SKEWNESS":           (19, get_skewness),
+    "KURTOSIS":           (20, get_kurtosis),
+    "MINIMUM":            (21, get_minimum),
+    "MAXIMUM":            (22, get_maximum),
+    "MEDIAN":             (23, get_median),
+    "PERCENTILE":         (24, get_percentile),
+    "HISTOGRAM":          (25, get_histogram)
+}
+
+
+modeItems = [
+    ("INT", "Int", "", "", 0),
+    ("FLOAT", "Float", "", "", 1)]
+
+functionItems = [(k, k.title(), "", "", s[0]) for k, s in sorted(functions.items(), key=lambda k: k[1][0])]
+
+
+class SvListStatisticsNode(bpy.types.Node, SverchCustomTreeNode):
+    '''
+    Triggers: Sum, Avg, Min, Max
+    Tooltip: Statistical quantities: sum, average, standard deviation, min, max, product...
+    '''
+    bl_idname = 'SvListStatisticsNode'
+    bl_label = 'List Statistics'
+    bl_icon = 'OUTLINER_OB_EMPTY'
+
+    def update_function(self, context):
+        if self.function == "ALL STATISTICS":
+            self.inputs["Percentage"].hide_safe = False
+            self.inputs["Bins"].hide_safe = False
+            self.inputs["Size"].hide_safe = not self.normalize
+            self.outputs[0].name = "Names"
+            self.outputs[1].name = "Values"
+        else:
+            for name in ["Percentage", "Bins", "Size"]:
+                self.inputs[name].hide_safe = True
+            if self.function == "PERCENTILE":
+                self.inputs["Percentage"].hide_safe = False
+            elif self.function == "HISTOGRAM":
+                self.inputs["Bins"].hide_safe = False
+                self.inputs["Size"].hide_safe = not self.normalize
+
+            self.outputs[0].name = "Name"
+            self.outputs[1].name = "Value"
+
+        updateNode(self, context)
+
+    def update_normalize(self, context):
+        socket = self.inputs["Size"]
+        socket.hide_safe = not self.normalize
+
+        updateNode(self, context)
+
+    mode = EnumProperty(
+        name="Mode", items=modeItems, default="FLOAT", update=updateNode)
+
+    function = EnumProperty(
+        name="Function", items=functionItems, update=update_function)
+
+    percentage = FloatProperty(
+        name="Percentage",
+        default=0.75, min=0.0, max=1.0, update=updateNode)
+
+    bins = IntProperty(
+        name="Bins",
+        default=10, min=1, update=updateNode)
+
+    normalize = BoolProperty(
+        name="Normalize", description="Normalize the bins to a normalize size",
+        default=False, update=update_normalize)
+
+    normalized_size = FloatProperty(
+        name="Size", description="The normalized size of the bins",
+        default=10.0, update=updateNode)
+
+    def draw_buttons(self, context, layout):
+        layout.prop(self, "mode", expand=True)
+        layout.prop(self, "function", text="")
+        if self.function in ["HISTOGRAM", "ALL STATISTICS"]:
+            layout.prop(self, "normalize", toggle=True)
+
+    def sv_init(self, context):
+        self.width = 150
+        self.inputs.new('StringsSocket', "Data")
+        self.inputs.new('StringsSocket', "Percentage").prop_name = "percentage"
+        self.inputs.new('StringsSocket', "Bins").prop_name = "bins"
+        self.inputs.new('StringsSocket', "Size").prop_name = "normalized_size"
+        self.outputs.new('StringsSocket', "Names")
+        self.outputs.new('StringsSocket', "Values")
+        self.function = "AVERAGE"
+
+    def get_statistics_function(self):
+        return functions[self.function][1]
+
+    def process(self):
+        outputs = self.outputs
+        # return if no outputs are connected
+        if not any(s.is_linked for s in outputs):
+            return
+
+        inputs = self.inputs
+        input_D = inputs["Data"].sv_get()
+        input_P = inputs["Percentage"].sv_get()[0]
+        input_B = inputs["Bins"].sv_get()[0]
+        input_S = inputs["Size"].sv_get()[0]
+
+        # sanitize the inputs
+        input_P = list(map(lambda x: max(0, min(1, x)), input_P))
+        input_B = list(map(lambda x: max(1, x), input_B))
+
+        if self.mode == "INT":
+            input_P = list(map(lambda x: int(x), input_P))
+
+        if self.function == "ALL STATISTICS":
+            functionNames = [fn[0] for fn in functionItems[1:]]
+        else:
+            functionNames = [self.function]
+
+        params = match_long_repeat([input_D, input_P, input_B, input_S])
+
+        allNames = []
+        allValues = []
+        for functionName in functionNames:
+            statistics_function = functions[functionName][1]
+            quantityList = []
+            for d, p, b, s in zip(*params):
+                if functionName == "PERCENTILE":
+                    quantity = statistics_function(d, p)
+                elif functionName == "HISTOGRAM":
+                    quantity = statistics_function(d, b, self.normalize, s)
+                else:
+                    quantity = statistics_function(d)
+
+                if functionName != "HISTOGRAM":
+                    if self.mode == "INT":
+                        quantity = int(quantity)
+
+                quantityList.append(quantity)
+
+            allNames.append(functionName)
+            allValues.append(quantityList)
+
+        outputs[0].sv_set(allNames)
+        outputs[1].sv_set(allValues)
+
+
+def register():
+    bpy.utils.register_class(SvListStatisticsNode)
+
+
+def unregister():
+    bpy.utils.unregister_class(SvListStatisticsNode)

utils/modules/statistics_functions.py

@@ -0,0 +1,121 @@
+# ##### BEGIN GPL LICENSE BLOCK #####
+#
+#  This program is free software; you can redistribute it and/or
+#  modify it under the terms of the GNU General Public License
+#  as published by the Free Software Foundation; either version 2
+#  of the License, or (at your option) any later version.
+#
+#  This program is distributed in the hope that it will be useful,
+#  but WITHOUT ANY WARRANTY; without even the implied warranty of
+#  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+#  GNU General Public License for more details.
+#
+#  You should have received a copy of the GNU General Public License
+#  along with this program; if not, write to the Free Software Foundation,
+#  Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
+#
+# ##### END GPL LICENSE BLOCK #####
+
+from functools import reduce
+from math import sqrt, floor
+import sys
+
+
+def get_sum(values):
+    return sum(values)
+
+
+def get_sum_of_squares(values):
+    return sum([v * v for v in values])
+
+
+def get_sum_of_inversions(values):
+    return sum([1.0 / v for v in values])
+
+
+def get_product(values):
+    return reduce((lambda x, y: x * y), values)
+
+
+def get_average(values):
+    return sum(values) / len(values)
+
+
+def get_geometric_mean(values):
+    return pow(get_product(values), 1.0 / len(values))
+
+
+def get_harmonic_mean(values):
+    return len(values) / get_sum_of_inversions(values)
+
+
+def get_standard_deviation(values):
+    a = get_average(values)
+    return sqrt(sum([(v - a)**2 for v in values]))
+
+
+def get_root_mean_square(values):
+    return sqrt(get_sum_of_squares(values) / len(values))
+
+
+def get_skewness(values):
+    a = get_average(values)
+    n = len(values)
+    s = get_standard_deviation(values)
+    return sum([(v - a)**3 for v in values]) / n / pow(s, 3)
+
+
+def get_kurtosis(values):
+    a = get_average(values)
+    n = len(values)
+    s = get_standard_deviation(values)
+    return sum([(v - a)**4 for v in values]) / n / pow(s, 4)
+
+
+def get_minimum(values):
+    return min(values)
+
+
+def get_maximum(values):
+    return max(values)
+
+
+def get_median(values):
+    sortedValues = sorted(values)
+    index = int(floor(len(values) / 2))
+    print("index=", index)
+    if len(values) % 2 == 0:  # even number of values ? => take the average of central values
+        median = (sortedValues[index - 1] + sortedValues[index]) / 2
+    else:  # odd number of values ? => take the central value
+        median = sortedValues[index]
+
+    return median
+
+
+def get_percentile(values, percentage):
+    sortedValues = sorted(values)
+    index = int(min(int(floor(len(values) * percentage)), len(values) - 1))
+    return sortedValues[index]
+
+
+def get_histogram(values, numBins, normalize=False, normalizedSize=10):
+    minValue = get_minimum(values)
+    maxValue = get_maximum(values)
+
+    binSize = max((maxValue - minValue) / numBins, sys.float_info.min)
+
+    # initialize the histogram bins
+    histogram = [0] * numBins
+
+    # populate the histogram bins
+    for i in range(len(values)):
+        binIndex = int(min(int(floor((values[i] - minValue) / binSize)), numBins - 1))
+        histogram[binIndex] = histogram[binIndex] + 1
+
+    # normalize histogram ?
+    if normalize:
+        binMax = max(histogram)
+        for i in range(len(histogram)):
+            histogram[i] = histogram[i] / binMax * normalizedSize
+
+    return histogram