Merge pull request #26 from mikkokotila/master

1.9.5 feature update

astetik/__init__.py

@@ -20,15 +20,33 @@
 from .plots.bartwo import bartwo
 from .plots.animate import Animation
 from .plots.world import world
+from .plots.regs import regs
 # from .plots.words import words
 
 
 # TABLES
-from .tables.ols import ols
 from .tables.table import table
 from .tables.text import text
 # from .tables.timeseries import timeseries
 
-del plots, style, tables, utils
+try:
+    del plots
+except:
+    pass
 
-__version__ = "1.9.3"
+try:
+    del style
+except:
+    pass
+
+try:
+    del tables
+except:
+    pass
+
+try:
+    del utils
+except:
+    pass
+
+__version__ = "1.9.4"

astetik/plots/__init__.py

@@ -1 +1 @@
-__version__ = "1.9.3"
+__version__ = "1.9.4"

astetik/plots/bar.py

@@ -37,9 +37,9 @@ def bar(data,
 
     1. USE
     ======
-    ast.bar1d(data=patients,
-              x='icu_days',
-              y='insurance')
+    ast.bar(data=patients,
+            x='icu_days',
+            y='insurance')
 
     2. PARAMETERS
     =============
@@ -122,7 +122,7 @@ def bar(data,
         data.columns = [y, x]
 
     if sort != None:
-        data = data.sort_values(x, ascending=sort)
+        data = data.sort_values(y, ascending=sort)
 
     if multi_color == True:
         n_colors = len(data[x].unique())

astetik/plots/hist.py

@@ -32,7 +32,9 @@ def hist(data,
 
     '''HISTOGRAM
 
-    Create a classic histogram.
+    Create a classic histogram. Note that nans will be
+    dropped automatically as otherwise seaborn will throw
+    an error.
 
     Inputs: 1
     Features: One continuous
@@ -123,7 +125,7 @@ def hist(data,
     p, ax = plt.subplots(figsize=(params()['fig_width'],
                                   params()['fig_height']))
 
-    p = sns.distplot(data[x],
+    p = sns.distplot(data[x].dropna(),
                      bins=bins,
                      norm_hist=False,
                      kde=False,

astetik/plots/regs.py

@@ -0,0 +1,147 @@
+import matplotlib.pyplot as plt
+import seaborn as sns
+
+from ..style.formats import _thousand_sep
+from ..style.style import params
+from ..style.titles import _titles
+from ..style.template import _header, _footer
+from ..utils.utils import _limiter, _scaler
+
+
+def regs(data,
+         x,
+         y,
+         hue=None,
+         marker='x',
+         fit_reg=False,
+         draw_scatter=True,
+         logres=False,
+         palette='default',
+         style='astetik',
+         dpi=72,
+         title='',
+         sub_title='',
+         x_label='',
+         y_label='',
+         legend=True,
+         x_scale='linear',
+         y_scale='linear',
+         x_limit=None,
+         y_limit=None,
+         outliers=False,
+         save=False):
+
+    '''REGRESSION SCATTER PLOT
+
+    This can be used as a standard two variable scatter plot
+    where both variables are continuous. If x is categorical,
+    use scat() instead or swarm(), depending on your need.
+    Here you can also choose logres as True in case one of your
+    variables is a binary categorical.
+
+    1. USE
+    ======
+    p = regs(data=df,
+         x='Age',
+         y='Fare',
+         palette='default',
+         style='astetik')
+
+    2. PARAMETERS
+    =============
+    2.1 INPUT PARAMETERS
+    --------------------
+    data :: pandas dataframe
+
+    x :: x-axis data
+
+    y :: y-axis data
+
+    --------------------
+    2.2. PLOT PARAMETERS
+    --------------------
+    marker :: By default is 'x' but you can use any matplotlib supported
+
+    fit_reg :: If the regression fit should be drawn
+
+    draw_scatter :: if the scatter should be drawn
+
+    logres :: for logistic regression when one feature is binary categorical.
+
+    ----------------------
+    2.3. COMMON PARAMETERS
+    ----------------------
+    palette :: One of the hand-crafted 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'
+
+    '''
+
+    # HEADER STARTS >>>
+    palette = _header(palette, style, n_colors=1, dpi=dpi)  # NOTE: y exception
+    # <<< HEADER ENDS
+
+    # # # # # # PLOT CODE STARTS # # # # # #
+    p, ax = plt.subplots(figsize=(params()['fig_width'],
+                                  params()['fig_height']))
+
+    sns.regplot(data=data,
+                x=x, y=y,
+                fit_reg=fit_reg,
+                scatter=draw_scatter,
+                color=palette[0],
+                marker=marker,
+                logistic=logres)
+    # # # # # # PLOT CODE ENDS # # # # # #
+
+    # SCALING AND LIMITS STARTS >>>
+    if x_scale != 'linear' or y_scale != 'linear':
+        _scaler(p, x_scale, y_scale)
+
+    if x_limit != None or y_limit != None:
+        _limiter(data=data, x=x, y=y, x_limit=x_limit, y_limit=y_limit)
+
+    # START OF TITLES >>>
+    _titles(title, sub_title=sub_title)
+    _thousand_sep(p, ax)
+    _footer(p=p, xlabel=x_label, ylabel=y_label, legend=False, n=1, save=save)
+
+    ax.xaxis.set_major_locator(plt.MaxNLocator(5))

astetik/plots/scat.py

@@ -31,14 +31,16 @@ def scat(data,
 
     '''SCATTER PLOT
 
-    Observations may overlap, and sizing is possible. All except 'hue' should
-    be continuous variables (or sizing can be stepped). If you want to use
-    categorical on x-axis, use ast.swarm() or ast.strip() instead.
+    Well actually, in comparison to a traditional scatter plot, there is
+    are some limimations here. Observations may overlap, and sizing is
+    possible. 'x' and 'hue' should not be continuous variables
+    ('size' can be stepped). If you want basic scatter plot, use regs() and
+    also see swarm() and strip() depending on your data.
 
     1. USE
     ======
     p = scat(data=df,
-         x='Age',
+         x='Class',
          y='Fare',
          hue='Survived',
          size='Rand',
@@ -149,7 +151,7 @@ def scat(data,
 
     # START OF TITLES >>>
     _titles(title, sub_title=sub_title)
-    _thousand_sep(p, ax)
+    #_thousand_sep(p, ax)
     _footer(p, x_label, y_label, legend, n, save)
 
-    ax.xaxis.set_major_locator(plt.MaxNLocator(5))
+    #ax.xaxis.set_major_locator(plt.MaxNLocator(5))

astetik/style/__init__.py

@@ -1 +1 @@
-__version__ = "1.9.3"
+__version__ = "1.9.4"

astetik/tables/__init__.py

@@ -1 +1 @@
-__version__ = "1.9.3"
+__version__ = "1.9.4"

astetik/utils/__init__.py

@@ -1 +1 @@
-__version__ = "1.9.3"
+__version__ = "1.9.4"

astetik/utils/country_code.py

@@ -2,7 +2,7 @@
 
 import pkg_resources
 
-countries = pkg_resources.resource_filename(__name__, '../extras/countries.csv')
+countries = pkg_resources.resource_filename('astetik', 'extras/countries.csv')
 countries = pd.read_csv(countries)
 
 

requirements.txt

@@ -1,8 +1,7 @@
-matplotlib
 seaborn
 geonamescache
-statsmodels
 numpy
 pandas
 IPython
 patsy
+statsmodels

setup.py

@@ -24,7 +24,7 @@
 URL = 'http://mikkokotila.com'
 LICENSE = 'MIT'
 DOWNLOAD_URL = 'https://github.com/mikkokotila/pretty'
-VERSION = '1.9.3'
+VERSION = '1.9.4'
 
 try:
     from setuptools import setup
@@ -35,6 +35,10 @@
 def check_dependencies():
     install_requires = []
 
+    try:
+        import matplotlib
+    except ImportError:
+        install_requires.append('matplotlib')
     try:
         import seaborn
     except ImportError:
@@ -43,10 +47,6 @@ def check_dependencies():
         import numpy
     except ImportError:
         install_requires.append('numpy')
-    #try:
-    #    import matplotlib
-    #except ImportError:
-    #    install_requires.append('matplotlib')
     try:
         import pandas
     except ImportError:
@@ -55,10 +55,6 @@ def check_dependencies():
         import IPython
     except ImportError:
         install_requires.append('IPython')
-    try:
-        import statsmodels
-    except ImportError:
-        install_requires.append('statsmodels')
     try:
         import patsy
     except ImportError:
@@ -67,7 +63,10 @@ def check_dependencies():
         import geonamescache
     except ImportError:
         install_requires.append('geonamescache')
-
+    try:
+        import statsmodels
+    except ImportError:
+        install_requires.append('statsmodels')
 
     return install_requires