[routines/heatamp] Incorporates unit scaling into x-axis ticks

Heatmap added x-ticks based on the time points, not absolute time.

When a DataFrame is passed into matplotlib.imshow, the Axes no longer
cares about the column names.  We want the plot to take into account the
sampling period AND the unit scaling if the user specifies one.

I copied over the method I implemented for boxplot.py -- it essentially
'tricks' matplotlib by redefining the labels rather than change the
time-axis values (see
https://stackoverflow.com/questions/10171618/changing-plot-scale-by-a-factor-in-matplotlib).
matplotlib's xscale does not support simple linear re-scaling; in any
case, having time-axis values does not make sense for imshow.

This commit addresses issue #20.

routines/heatmap.py

@@ -2,7 +2,7 @@
 
 import numpy as np
 import matplotlib.pyplot as plt
-from matplotlib import cm, colors
+from matplotlib import cm, colors, ticker
 
 from postprocessor.core.processes.standardscaler import standardscaler
 from postprocessor.routines.plottingabc import BasePlotter
@@ -17,14 +17,14 @@ class _HeatmapPlotter(BasePlotter):
         trace_name,
         births_df,
         cmap,
-        sampling_period,
+        unit_scaling,
         xtick_step,
         scale,
         robust,
         xlabel,
         plot_title,
     ):
-        super().__init__(trace_name, sampling_period, xlabel, plot_title)
+        super().__init__(trace_name, unit_scaling, xlabel, plot_title)
         # Define attributes from arguments
         self.trace_df = trace_df
         self.births_df = births_df
@@ -59,12 +59,17 @@ class _HeatmapPlotter(BasePlotter):
             self.vmin = None
             self.vmax = None
 
+        # Define horizontal axis ticks and labels
+        # hacky! -- redefine column names
+        trace_df.columns = trace_df.columns * self.unit_scaling
+        self.fmt = ticker.FuncFormatter(
+            lambda x, pos: "{0:g}".format(x * self.unit_scaling)
+        )
+
     def plot(self, ax, cax):
         """Draw the heatmap on the provided Axes."""
         super().plot(ax)
-        # Horizontal axis labels as multiples of xtick_step
-        ax.xaxis.set_major_locator(plt.MultipleLocator(self.xtick_step))
-
+        ax.xaxis.set_major_formatter(self.fmt)
         # Draw trace heatmap
         trace_heatmap = ax.imshow(
             self.trace_scaled,
@@ -73,7 +78,11 @@ class _HeatmapPlotter(BasePlotter):
             vmin=self.vmin,
             vmax=self.vmax,
         )
-
+        # Horizontal axis labels as multiples of xtick_step, taking
+        # into account unit scaling
+        ax.xaxis.set_major_locator(
+            ticker.MultipleLocator(self.xtick_step / self.unit_scaling)
+        )
         # Overlay births, if present
         if self.births_df is not None:
             # Must be masked array for transparency
@@ -86,7 +95,6 @@ class _HeatmapPlotter(BasePlotter):
                 births_heatmap_mask,
                 interpolation="none",
             )
-
         # Draw colour bar
         colorbar = ax.figure.colorbar(
             mappable=trace_heatmap, cax=cax, ax=ax, label=self.colorbarlabel
@@ -98,7 +106,7 @@ def heatmap(
     trace_name,
     births_df=None,
     cmap=cm.RdBu,
-    sampling_period=5,
+    unit_scaling=1,
     xtick_step=60,
     scale=True,
     robust=True,
@@ -120,8 +128,8 @@ def heatmap(
         0 or 1.
     cmap : matplotlib ColorMap
         Colour map for heatmap.
-    sampling_period : int or float
-        Sampling period, in unit time.
+    unit_scaling : int or float
+        Unit scaling factor, e.g. 1/60 to convert minutes to hours.
     xtick_step : int or float
         Interval length, in unit time, to draw x axis ticks.
     scale : bool
@@ -154,7 +162,7 @@ def heatmap(
         trace_name,
         births_df,
         cmap,
-        sampling_period,
+        unit_scaling,
         xtick_step,
         scale,
         robust,