Merge pull request #12 from phobson/cleanup-imports

Cleanup imports

Author: phobson

README.md

@@ -4,6 +4,15 @@ Real probability scales for matplotlib
 [![Build Status](https://travis-ci.org/phobson/mpl-probscale.svg)](https://travis-ci.org/phobson/mpl-probscale)
 [![Coverage Status](https://coveralls.io/repos/phobson/mpl-probscale/badge.svg?branch=master&service=github)](https://coveralls.io/github/phobson/mpl-probscale?branch=master)
 
+[Sphinx Docs](http://phobson.github.io/mpl-probscale/)
+
+## Installation
+
+`conda install mpl-probscale --channel=phobson`
+
+
+## Quick start
+
 Simply importing `probscale` lets you use probability scales in your matplotlib figures:
 
 ```python

docs/index.rst

@@ -15,6 +15,15 @@ mpl-probscale: Real probability scales for matplotlib
 
 https://github.com/phobson/mpl-probscale
 
+Installation
+------------
+
+``conda install mpl-probscale --channel=phobson``
+
+
+Quickstart
+----------
+
 Simply importing ``probscale`` lets you use probability scales in your matplotlib figures:
 
 .. code-block:: python

docs/tutorial/getting_started.ipynb

@@ -178,19 +178,18 @@
    "cell_type": "code",
    "execution_count": null,
    "metadata": {
-    "collapsed": false,
-    "scrolled": false
+    "collapsed": false
    },
    "outputs": [],
    "source": [
     "import paramnormal\n",
     "seaborn.set(style='ticks', context='notebook', rc=clear_bkgd)\n",
     "\n",
-    "fig, (ax1, ax2, ax3, ax4) = pyplot.subplots(figsize=(8, 5), nrows=4)\n",
+    "fig, (ax1, ax2, ax3, ax4) = pyplot.subplots(figsize=(8, 6), nrows=4)\n",
     "\n",
     "ax1.set_xscale('prob')\n",
     "ax1.set_xlim(left=2, right=98)\n",
-    "ax1.set_xlabel('Normal probability scale')\n",
+    "ax1.set_xlabel('Normal probability scale, as percents')\n",
     "ax1.set_yticks([])\n",
     "\n",
     "beta1 = paramnormal.beta(α=3, β=2)\n",
@@ -206,7 +205,7 @@
     "ax3.set_yticks([])\n",
     "\n",
     "ax4.set_xlim(left=2, right=98)\n",
-    "ax4.set_xticks(ax1.get_xticks()[14:-14])\n",
+    "ax4.set_xticks(ax1.get_xticks()[12:-12])\n",
     "ax4.set_xlabel('Linear scale (for reference)')\n",
     "ax4.set_yticks([])\n",
     "\n",
@@ -354,4 +353,4 @@
  },
  "nbformat": 4,
  "nbformat_minor": 0
-}
+}

probscale/__init__.py

@@ -1 +1,6 @@
+from matplotlib import scale
+
 from .viz import *
+from .probscale import ProbScale
+
+scale.register_scale(ProbScale)

probscale/formatters.py

@@ -0,0 +1,96 @@
+import numpy as numpy
+from matplotlib.ticker import Formatter
+
+
+class _FormatterMixin(Formatter):
+    @classmethod
+    def _sig_figs(cls, x, n, expthresh=5, forceint=False):
+        """ Formats a number with the correct number of sig figs.
+
+        Parameters
+        ----------
+        x : int or float
+            The number you want to format.
+        n : int
+            The number of significan figures it should have.
+        expthresh : int, optional (default = 5)
+            The absolute value of the order of magnitude at which numbers
+            are formatted in exponential notation.
+        forceint : bool, optional (default is False)
+            If true, simply returns int(x)
+
+        Returns
+        -------
+        formatted : str
+            The formatted number as a string
+
+        Examples
+        --------
+        >>> print(_sig_figs(1247.15, 3))
+               1250
+        >>> print(_sig_figs(1247.15, 7))
+               1247.150
+
+        """
+
+        # return a string value unaltered
+        if isinstance(x, str) or x == 0.0:
+            out = str(x)
+
+        # check on the number provided
+        elif x is not None and not numpy.isinf(x) and not numpy.isnan(x):
+
+            # check on the _sig_figs
+            if n < 1:
+                raise ValueError("number of sig figs (n) must be greater than zero")
+
+            elif forceint:
+                out = '{:,.0f}'.format(x)
+
+            # logic to do all of the rounding
+            else:
+                order = numpy.floor(numpy.log10(numpy.abs(x)))
+
+                if (-1.0 * expthresh <= order <= expthresh):
+                    decimal_places = int(n - 1 - order)
+
+                    if decimal_places <= 0:
+                        out = '{0:,.0f}'.format(round(x, decimal_places))
+
+                    else:
+                        fmt = '{0:,.%df}' % decimal_places
+                        out = fmt.format(x)
+
+                else:
+                    decimal_places = n - 1
+                    fmt = '{0:.%de}' % decimal_places
+                    out = fmt.format(x)
+
+        # with NAs and INFs, just return 'NA'
+        else:
+            out = 'NA'
+
+        return out
+
+    def __call__(self, x, pos=None):
+        if x < (10 / self.factor):
+            out = self._sig_figs(x, 1)
+        elif x <= (99 / self.factor):
+            out =  self._sig_figs(x, 2)
+        else:
+            order = numpy.ceil(numpy.round(numpy.abs(numpy.log10(self.top - x)), 6))
+            out = self._sig_figs(x, order + self.offset)
+
+        return '{}'.format(out)
+
+
+class PctFormatter(_FormatterMixin):
+    factor = 1.0
+    offset = 2
+    top = 100
+
+
+class ProbFormatter(_FormatterMixin):
+    factor = 100.0
+    offset = 0
+    top = 1