release:grplot-0.11

grplot/__init__.py

@@ -10,7 +10,7 @@
 from grplot.utils.strtoarray import strtoarray
 
 
-def grplot(plot, # default general value
+def plot2d(plot, # default general value
            df, 
            x=None, 
            y=None, 
@@ -195,14 +195,14 @@ def grplot(plot, # default general value
     by ghiffary rifqialdi
     based on numpy, scipy, matplotlib, seaborn, squarify, and pandas
     
-    version = '0.10.4'
+    version = '0.11'
 
     release date
-    25/08/2022
+    11/09/2022
     -----------------------------------------------
 
     documentation is available at https://github.com/ghiffaryr/grplot
-    '''    
+    '''
     # initialization
     # creating figure
     Nx, Ny = check_axes(x, y, Nx, Ny)

grplot/analytic/__init__.py

@@ -0,0 +1,136 @@
+import numpy as np
+import pandas as pd
+import grplot_seaborn as sns
+import matplotlib
+from matplotlib import pyplot as plt
+import matplotlib.dates as mdates
+from grplot.features.optimizer.optimizer_data import optimizer_data
+from grplot.features.saveas.check_saveas import check_saveas
+from grplot.features.sep.statdesc_sep.statdesc_sep_def import statdesc_sep_def
+
+
+def cohort(df,
+           signup_date, 
+           last_active_date, 
+           figsize=[8,6],
+           fontsize=10, 
+           tick_fontsize=None, 
+           legend_fontsize=None, 
+           text_fontsize=None, 
+           label_fontsize=None, 
+           title_fontsize=None, 
+           sep=',',
+           saveas=None,
+           display_summary=False):
+    '''
+    -----------------------------------------------
+    grplot: lazy statistical data visualization
+    
+    by ghiffary rifqialdi
+    based on numpy, scipy, matplotlib, seaborn, squarify, and pandas
+    
+    version = '0.11'
+
+    release date
+    11/09/2022
+    -----------------------------------------------
+
+    documentation is available at https://github.com/ghiffaryr/grplot
+    '''
+    if type(signup_date) == str and type(last_active_date) == str:
+        # preprocessing
+        preprocessing = optimizer_data(plot='heatmap', 
+                                       df=df, 
+                                       x=signup_date, 
+                                       y=last_active_date, 
+                                       hue=None, 
+                                       size=None, 
+                                       style=None, 
+                                       units=None, 
+                                       axes=None, 
+                                       mode='perf') \
+                        .copy(deep=True)
+        preprocessing['Customer ID'] = np.arange(len(preprocessing)) + 1
+        preprocessing['Signup Date'] = pd.to_datetime(preprocessing[signup_date]).dt.tz_localize(None).dt.to_period('M').dt.to_timestamp()
+        preprocessing['Last Active Date'] = pd.to_datetime(preprocessing[last_active_date]).dt.tz_localize(None).dt.to_period('M').dt.to_timestamp()
+        preprocessing = preprocessing[['Customer ID', 'Signup Date', 'Last Active Date']]
+        preprocessing.set_index('Customer ID', inplace=True)
+        preprocessing['Metric'] = [pd.period_range(s, e, freq='m') for s, e in zip(preprocessing['Signup Date'], preprocessing['Last Active Date'])]
+        preprocessing = preprocessing.explode('Metric')
+        preprocessing['Cohort Group'] = preprocessing.groupby(level=0)['Metric'].min().dt.strftime('%Y-%m')
+        preprocessing.reset_index(inplace=True)
+        cohort = preprocessing.groupby(['Cohort Group', 'Metric']).agg({'Customer ID':pd.Series.nunique})
+        cohort.rename(columns={'Customer ID': 'Total Account'}, inplace=True)    
+        def cohort_period(df):
+            df['Cohort Period'] = np.arange(len(df))
+            return df
+        cohort = cohort.groupby(level=0).apply(cohort_period)
+        cohort.reset_index(inplace=True)
+        cohort.set_index(['Cohort Group', 'Cohort Period'], inplace=True)
+        # summary
+        if display_summary == True: 
+            display(cohort)
+        else:
+            pass
+        cohort_group_size = cohort['Total Account'].groupby(level=0).first()
+        user_retention = cohort['Total Account'].unstack(0).divide(cohort_group_size, axis=1)
+        user_retention = user_retention.iloc[0:,:len(user_retention)]
+        user_retention = user_retention * 100
+        # plotting
+        # creating figure
+        fig, ax = plt.subplots(figsize=(figsize[0], 
+                                        figsize[1]))
+        # check fontsize
+        if tick_fontsize is None:
+            tick_fontsize = fontsize
+        else:
+            pass
+        if legend_fontsize is None:
+            legend_fontsize = fontsize
+        else:
+            pass
+        if text_fontsize is None:
+            text_fontsize = fontsize
+        else:
+            pass
+        if label_fontsize is None:
+            label_fontsize = fontsize
+        else:
+            pass
+        if title_fontsize is None:
+            title_fontsize = fontsize
+        else:
+            pass
+        sns.heatmap(user_retention.T, mask=user_retention.T.isnull(), annot=True,  annot_kws={"size":text_fontsize}, fmt='.2f', cmap='RdYlGn', xticklabels=list(range(0, len(user_retention))), ax=ax)
+        # fontsize
+        ax.set_title(label="Monthly Retention Rate", fontsize=title_fontsize)
+        ax.set_xlabel('Periods in Month', fontsize=label_fontsize)
+        ax.set_ylabel('Cohort Group (Signup Month)', fontsize=label_fontsize)
+        ax.tick_params(axis='both', labelsize=tick_fontsize)
+        cbar = ax.collections[0].colorbar
+        cbar.ax.tick_params(labelsize=legend_fontsize)
+        # rot
+        for label in ax.get_yticklabels():
+            label.set_rotation(0)
+        # sep
+        # sep cbar
+        yticks = []
+        for num in cbar.ax.get_yticks():
+            num = round(float(num), 2)
+            num = statdesc_sep_def(num, sep)
+            yticks.append(num + '%')
+        cbar.ax.set_yticks(cbar.ax.get_yticks())
+        cbar.ax.set_yticklabels(yticks)
+        # sep text
+        for child in ax.get_children():
+            if isinstance(child, matplotlib.text.Text) and (child.get_text().replace('.', '', 1).isdigit() == True):
+                num = float(child.get_text())
+                num = statdesc_sep_def(num, sep)
+                child.set_text(num + '%')
+            else:
+                pass
+        # save image as
+        check_saveas(fig, saveas)
+    else:
+        raise Exception('Wrong data type of axis!')
+    return ax

grplot/features/plot/plot_single_def.py

@@ -1,5 +1,5 @@
 import numpy
-import seaborn as sns
+import grplot_seaborn as sns
 from matplotlib.ticker import PercentFormatter
 from pandas.api.types import is_numeric_dtype, is_object_dtype, is_categorical_dtype
 from grplot.features.plot.packedbubbles import plot as pb

grplot/features/sep/text_sep/text_sep_type.py

@@ -1,4 +1,4 @@
-from pandas.api.types import is_numeric_dtype
+from pandas.api.types import is_float_dtype
 from grplot.features.sep.text_sep.text_sep_data_def import text_sep_data_def
 from grplot.utils.arg_ax_type import arg_ax_type
 from grplot.utils.arg_axis_ax_type import arg_axis_ax_type
@@ -10,7 +10,7 @@ def text_sep_type(plot, df, num, sep, axislabel, axes):
     else:
         sep = arg_axis_ax_type(arg=sep, axislabel=axislabel, axes=axes)
     if sep is None:
-        if is_numeric_dtype(type(num)) == True: 
+        if is_float_dtype(type(num)) == True: 
             if num.is_integer() == True:
                 num = '{}'.format(int(num))
             else:

grplot_seaborn/__init__.py

@@ -0,0 +1,21 @@
+# import grplot_seaborn objects
+from .rcmod import *  # noqa: F401,F403
+from .utils import *  # noqa: F401,F403
+from .palettes import *  # noqa: F401,F403
+from .relational import *  # noqa: F401,F403
+from .regression import *  # noqa: F401,F403
+from .categorical import *  # noqa: F401,F403
+from .distributions import *  # noqa: F401,F403
+from .matrix import *  # noqa: F401,F403
+from .miscplot import *  # noqa: F401,F403
+from .axisgrid import *  # noqa: F401,F403
+from .widgets import *  # noqa: F401,F403
+from .colors import xkcd_rgb, crayons  # noqa: F401
+from . import cm  # noqa: F401
+
+# Capture the original matplotlib rcParams
+import matplotlib as mpl
+_orig_rc_params = mpl.rcParams.copy()
+
+# Define the seaborn version
+__version__ = "0.11.2"