getting ready for 2.0 release

- added Animator
- added support for pie chart into animator
- added bar1 which is just for showing 1d data
- changed bars() slightly
- removed rotation from corr() labels
- added sort to count()
- many improvements to line()
- some improvements to overlap()
- pie() is more or less redone from ground up
- scat() is cleaned up
- added microseconds to the end of the saved file name string
- title is now produced only if there is input for title (no empty space in some plots)
- changed the way the tick params for timeseries are handled
- added one new exception for the cases where labels are missing
- added function for loading data (just placeholder for now)
- cleaned up and fixed some issues from transforms.py
- added a rational way for sizing of a single seaborn factorplot
- added random color option for colors
- added new 'colorblind' default that support several kinds of colorblindness
- added one more new colorblind palette with 14 colors

LOTS OF TESTING! :)

astetik/__init__.py

@@ -16,6 +16,8 @@
 from .plots.multikde import multikde
 from .plots.compare import compare
 from .plots.multicount import multicount
+from .plots.bar1 import bar1
+from .plots.animate import Animation
 # from .plots.words import words
 
 
@@ -25,4 +27,6 @@
 from .tables.text import text
 # from .tables.timeseries import timeseries
 
-__version__ = "1.1"
+from .utils.transform import boolcols_to_cat
+
+__version__ = "1.9"

astetik/plots/animate.py

@@ -0,0 +1,60 @@
+import glob
+import os
+import numpy as np
+import matplotlib.pyplot as plt
+
+from IPython.display import clear_output
+
+
+class Animation():
+
+    '''ANIMATED CHARTS
+
+    Supports dataformat where x and y are compared against each other, and
+    label_col provides the title (e.g. year) for each frame in the animation.
+
+    y and x are used for the legend labels when relevant.
+
+    '''
+
+    def __init__(self, data, x, y, label_col, plot_type, filename='animation',
+                 dpi=72, palette='default'):
+
+        self.plot_type = plot_type
+        self.data = data
+        self.x = x
+        self.y = y
+        self.label_col = label_col
+        self.palette = palette
+        self.filename = filename
+        self.frames = len(data)
+        self.dpi = dpi
+
+        # RUNTIME
+        self._ = self._create_plots()
+        self._ = self._create_gif()
+
+    def _create_plots(self):
+
+        for i in range(self.frames):
+            data = np.array([self.data[self.x][i], self.data[self.y][i]]).astype(int)
+            self.plot_type(data,
+                           labels=[self.x, self.y],
+                           palette=self.palette,
+                           sub_title=self.data[self.label_col][i],
+                           save=True,
+                           dpi=self.dpi)
+            clear_output()
+            plt.show()
+
+    def _create_gif(self):
+
+        gif_name = self.filename
+        file_list = glob.glob('astetik_*.png')
+        list.sort(file_list, key=lambda x: int(x.split('_')[1].split('.png')[0]))
+
+        with open('image_list.txt', 'w') as file:
+            for item in file_list:
+                file.write("%s\n" % item)
+
+        os.system('convert @image_list.txt {}.gif'.format(gif_name)) # magick on windows

astetik/plots/bar1.py

@@ -0,0 +1,140 @@
+# EXCEPTIONAL IMPORT #
+import matplotlib
+matplotlib.use('Agg')
+# ENDS #
+
+import matplotlib.pyplot as plt
+from matplotlib.pyplot import rcParams
+import seaborn as sns
+
+from ..style.template import _header, _footer
+from ..utils.utils import _limiter, _scaler
+from ..utils.utils import factorplot_sizing
+from ..style.titles import _titles
+
+
+def bar1(data,
+         x,
+         y,
+         multi_color=False,
+         palette='default',
+         style='astetik',
+         dpi=72,
+         title='',
+         sub_title='',
+         x_label='',
+         y_label='',
+         legend=True,
+         x_scale='linear',
+         y_scale='linear',
+         x_limit='auto',
+         y_limit='auto',
+         save=False):
+
+    '''BAR PLOT
+
+    A 1-dimensional bar graph for the case where there is a single
+    value per label.
+
+    Inputs: 2
+
+    1. USE
+    ======
+    ast.bar1d(data=patients,
+              x='icu_days',
+              y='insurance')
+
+    2. PARAMETERS
+    =============
+    2.1 INPUT PARAMETERS
+    --------------------
+    data :: pandas dataframe
+
+    x :: x-axis data (single value per label)
+
+    y :: y-axis data (labels)
+
+    --------------------
+    2.2. PLOT PARAMETERS
+    --------------------
+    multi_color :: If True, label values will be used for hue.
+
+    ----------------------
+    2.3. COMMON PARAMETERS
+    ----------------------
+    palette :: One of the astetik palettes:
+                'default'
+                'colorblind'
+                'blue_to_red'
+                'blue_to_green'
+                'red_to_green'
+                'green_to_red'
+                'violet_to_blue'
+                'brown_to_green'
+                'green_to_marine'
+
+                Or use any cmap, seaborn or matplotlib
+                color or palette code, or hex value.
+
+    style :: Use one of the three core styles:
+                'astetik'     # white
+                '538'         # grey
+                'solarized'   # sepia
+
+              Or alternatively use any matplotlib or seaborn
+              style definition.
+
+    dpi :: the resolution of the plot (int value)
+
+    title :: the title of the plot (string value)
+
+    sub_title :: a secondary title to be shown below the title
+
+    x_label :: string value for x-axis label
+
+    y_label :: string value for y-axis label
+
+    x_scale :: 'linear' or 'log' or 'symlog'
+
+    y_scale :: 'linear' or 'log' or 'symlog'
+
+    x_limit :: int or list with two ints
+
+    y_limit :: int or list with two ints
+
+    outliers :: Remove outliers using either 'zscore' or 'iqr'
+    '''
+
+    size, aspect = factorplot_sizing(data[x])
+
+    if multi_color == True:
+        n_colors = len(data[x].unique())
+    else:
+        n_colors = 1
+
+    # HEADER STARTS >>>
+    palette = _header(palette,
+                      style,
+                      n_colors=n_colors,
+                      dpi=dpi,
+                      fig_height=None,
+                      fig_width=None)
+    # <<< HEADER ENDS
+    p = sns.factorplot(data=data,
+                       x=x,
+                       y=y,
+                       palette=palette,
+                       aspect=aspect,
+                       size=size,
+                       kind='bar')
+
+    # SCALING AND LIMITS STARTS >>>
+    if x_scale != 'linear' or y_scale != 'linear':
+        _scaler(p, x_scale, y_scale)
+
+    # FOOTER STARTS >>>
+    _titles(title, sub_title=sub_title)
+    _footer(p, x_label, y_label, save=save)
+
+    if data[x].min() < 0:
+        sns.despine(left=True)

astetik/plots/bars.py

@@ -37,7 +37,7 @@ def bars(data,
     A multi-dimension bar plot that takes up to 5 features at a time.
 
     Inputs: 2 to 5
-    Features: At least one continuous (or stepped) variable and results
+    Features: At least one continuous (or stepped) variable and rest
               can be categorical.
 
     1. USE
@@ -116,6 +116,8 @@ def bars(data,
     outliers :: Remove outliers using either 'zscore' or 'iqr'
     '''
 
+    aspect = int(len(data[x].unique()) / 5)
+
     if hue != None:
         n_colors = len(data[hue].unique())
     else:
@@ -137,6 +139,7 @@ def bars(data,
                        col=col,
                        col_wrap=col_wrap,
                        palette=palette,
+                       aspect=aspect,
                        size=4,
                        kind='bar')
 

astetik/plots/corr.py

@@ -133,4 +133,4 @@ def corr(data,
     _titles(title, sub_title=sub_title)
     _footer(p, x_label, y_label, save=save)
 
-    p.set_xticklabels(data, rotation=45)
+    p.set_xticklabels(data)

astetik/plots/count.py

@@ -7,6 +7,7 @@
 
 def count(data,
           x,
+          sort=None,
           palette='default',
           style='astetik',
           dpi=72,
@@ -44,7 +45,7 @@ def count(data,
     --------------------
     2.2. PLOT PARAMETERS
     --------------------
-    None
+    sort :: If True, will be sorted based on input values.
 
     ----------------------
     2.3. COMMON PARAMETERS
@@ -95,6 +96,9 @@ def count(data,
 
     aspect = int(len(data[x].unique()) / 5)
 
+    if sort != None:
+        sort = data[x].value_counts().index.values
+
     # HEADER STARTS >>>
     palette = _header(palette,
                       style,
@@ -104,17 +108,18 @@ def count(data,
                       fig_width=None)
 
     p = sns.factorplot(data=data,
-                       x=x,
+                       y=x,
                        palette=palette,
                        size=4,
                        aspect=aspect,
                        kind='count',
                        legend=legend,
-                       legend_out=False)
+                       legend_out=False,
+                       order=sort)
 
     # FOOTER
     _titles(title, sub_title)
-    _thousand_sep(p, p.ax)
+    _thousand_sep(p, p.ax, y_sep=False)
     _footer(p, x_label, y_label, save=save)
 
     p.set_xticklabels(None, rotation=90)

astetik/plots/line.py

@@ -1,6 +1,7 @@
 import numpy as np
+import pandas as pd
 import matplotlib.pyplot as plt
-from matplotlib.dates import HourLocator
+from matplotlib import ticker
 
 from ..style.formats import _thousand_sep
 from ..style.style import params
@@ -164,7 +165,8 @@ def line(data,
                                   func=interval_func,
                                   freq=interval)
 
-    markers = ["o", "+", "x", "|", "-", ",", ".", "^", "v"]
+    markers = ["o", "+", "x", "|", "1", "8", "s", "p",
+               "o", "+", "x", "|", "1", "8", "s", "p"]
     # <<< END OF PLOT SPECIFIC
 
     # START OF HEADER >>>
@@ -205,7 +207,8 @@ def line(data,
 
     # DATETIME FORMAT
     if time_frame != None:
-        date_handler(data[x[0]], ax, time_frame)
+        data[y] = pd.to_datetime(data[y])
+        date_handler(data[y], ax, time_frame)
 
     # LIMITS
     if x_limit != None or y_limit != None:
@@ -217,7 +220,6 @@ def line(data,
     _footer(p, x_label, y_label, save=save)
 
     if legend != False:
-        plt.legend(x, loc=1, ncol=lines)
-    if y != None:
-        hours = HourLocator()
-        ax.xaxis.set_minor_locator(hours)
+        plt.legend(x, loc=1, ncol=1, bbox_to_anchor=(1.15, 1.0))
+
+    ax.xaxis.set_major_locator(ticker.MaxNLocator(nbins=8))

astetik/plots/overlap.py

@@ -1,5 +1,6 @@
 import seaborn as sns
 import matplotlib.pyplot as plt
+import matplotlib.patches as mpatches
 
 from ..style.titles import _titles
 from ..style.template import _header
@@ -12,6 +13,7 @@ def overlap(data,
             y,
             label_col,
             sort=None,
+            limit=None,
             transform_func=False,
             palette='default',
             style='astetik',
@@ -67,6 +69,8 @@ def overlap(data,
     sort :: either True or False for ascending sort based on the
             x-axis data.
 
+    limit :: limit the number of items to be shown
+
     transform_func :: If not False, the selected function such as
                       'mean' will be used to group by the label_col.
                       Available functions:
@@ -151,19 +155,25 @@ def overlap(data,
     sns.barplot(data=data,
                 x=x,
                 y=label_col,
+                orient='h',
                 color=palette[0])
 
     sns.barplot(data=data,
                 x=y,
                 y=label_col,
+                orient='h',
                 color=palette[1])
 
     # # # # PLOT ENDS # # # #
+    if legend != False:
+        x_patch = mpatches.Patch(color=palette[0], label=x)
+        y_patch = mpatches.Patch(color=palette[1], label=y)
+        ax.legend(handles=[x_patch, y_patch], ncol=1, loc="upper right", frameon=True)
 
-    ax.legend(ncol=2, loc="lower right", frameon=True)
     ax.set(ylabel=y_label, xlabel=x_label)
     sns.despine(bottom=True)
     ax.xaxis.set_major_locator(plt.MaxNLocator(5))
 
     _thousand_sep(p, ax)
-    _titles(title, sub_title=sub_title)
+    if len(title) + len(sub_title) < 0:
+        _titles(title, sub_title=sub_title)