18773402293

Committed 22 Oct 2025 07:27AM UTC coverage: 86.514% (-3.6%) from 90.141%

Build # 18773402293

Build Type

push

github

Committed by

olivergrasl

Commit Message

remove global

Run Details

6242 of 7215 relevant lines covered (86.51%)

0.87 hits per line

Source File
Press 'n' to go to next uncovered line, 'b' for previous

98.44

/BPTK_Py/visualizations/visualize.py

#                                                       /`-
# _                                  _   _             /####`-
# | |                                | | (_)           /########`-
# | |_ _ __ __ _ _ __  ___  ___ _ __ | |_ _ ___       /###########`-
# | __| '__/ _` | '_ \/ __|/ _ \ '_ \| __| / __|   ____ -###########/
# | |_| | | (_| | | | \__ \  __/ | | | |_| \__ \  |    | `-#######/
# \__|_|  \__,_|_| |_|___/\___|_| |_|\__|_|___/  |____|    `- # /
#
# Copyright (c) 2018 transentis labs GmbH
# MIT License



import statistics
import pandas as pd


class visualizer():
    """
    Class for building plots from dataframes. Includes capabilities to produce time series data.
    Can also only modify dataframes to generate time series and return the modified dataframe
    """

    def __init__(self,config=None):
        self.config = config

    def plot(self, df, return_df, visualize_from_period, visualize_to_period, stacked, kind, title, alpha, x_label,
             y_label, start_date="1/1/2018", freq="D", series_names={},format="plot"):
        """
        Plot method. Creates plots from dataframes
        :param df: DataFrame input
        :param return_df: Flag. If true return a dataFrame and do not plot (default: False)
        :param visualize_from_period: visualize from a specific t (default: 0)
        :param visualize_to_period:  visualize until a specific t (default: model's stoptime)
        :param stacked: If True, use stacked series (default: False)
        :param kind: 'area' or 'line' plot (ldefault: area)
        :param title: Title of plot
        :param alpha: Alpha of series (default: see config!)
        :param x_label: x_label of plot
        :param y_label: y_label of plot
        :param start_date: Start date for time series
        :param freq: Frequency setting for time series
        :param series_names: series renaming patterns
        :return: depends on format flag: either just a plot, which formally returns nothing, or Matplotlib axes, or a dataframe
        """

        if not kind:
            kind=self.config.configuration["kind"]

        if not stacked:
            stacked = self.config.configuration["stacked"]

        if not alpha:
            alpha = self.config.configuration["alpha"]

        if not start_date == "":
            df.index = pd.date_range(start_date, periods=len(df), freq=freq)

        series_names_keys = series_names.keys()

        if len(series_names) > 0:


            new_columns = {}
            for column in df.columns:
                for series_names_key in series_names_keys:
                    if series_names_key in column:
                        new_column = column.replace(series_names_key, series_names[series_names_key])
                        new_columns[column] = new_column

            df.rename(columns=new_columns, inplace=True)

        ## If user did not set return_df=True, plot the simulation results (default behavior)
        if not (return_df or format=="df"):

            ### Get the plot object
            if visualize_to_period == 0:

                ax = df.iloc[visualize_from_period:].plot(kind=kind, stacked=stacked,
                                                          figsize=self.config.configuration["figsize"],
                                                          title=title,
                                                          alpha=alpha, color=self.config.configuration["colors"],
                                                          lw=self.config.configuration["linewidth"])

            elif visualize_from_period == visualize_to_period:
                print("[INFO] No data to plot for period t={} to t={}".format(str(visualize_from_period),
                                                                              str(visualize_to_period)))
                return None

            else:
                if visualize_to_period + 1 > len(df):
                    visualize_to_period = len(df)

                ax = df.iloc[visualize_from_period:visualize_to_period].plot(kind=kind, stacked=stacked,
                                                                             figsize=self.config.configuration["figsize"],
                                                                             title=title,
                                                                             alpha=alpha,
                                                                             color=self.config.configuration["colors"],
                                                                             lw=self.config.configuration["linewidth"])
                ### Set axes labels and set the formats
            if (len(x_label) > 0):
                ax.set_xlabel(x_label)

                # Set the y-axis label
            if (len(y_label) > 0):
                ax.set_ylabel(y_label)

            for ymaj in ax.yaxis.get_majorticklocs():
                ax.axhline(y=ymaj, ls='-', alpha=0.05, color=(34.1 / 100, 32.9 / 100, 34.1 / 100))

            self.update_plot_formats(ax)

            if format=="axes":
                return ax
            else:
                return

        ### If user wanted a dataframe instead, here it is!
        elif return_df or format=="df":
            if visualize_to_period == 0:
                return df.iloc[visualize_from_period:]
            elif visualize_from_period == visualize_to_period:
                print("[INFO] No data for period t={} to t={}".format(str(visualize_from_period + 1),
                                                                      str(visualize_to_period + 1)))
                return None
            else:
                return df.iloc[visualize_from_period:visualize_to_period]



    def update_plot_formats(self, ax):
        """
        Configure the plot formats for the labels. Generates the formatting for y labels
        :param ax:
        :return:
        """
        ylabels_mean = statistics.mean(ax.get_yticks())


        # Override the format based on the mean values

        import matplotlib.ticker as ticker

        def label_format(x,pos):
            if ylabels_mean <= 2.0 and ylabels_mean >= -2.0:
                label = format(x, ',.2f')

            elif ylabels_mean <= 10.0 and ylabels_mean >= -10.0:
                label = format(x, ',.1f')

            else:
                label = format(x, ',.0f')

            return label


        ax.yaxis.set_major_formatter(ticker.FuncFormatter(func=label_format))

        ## Quick fix for incompatibility of scientific notation and ticklabels
        try:
            ax.ticklabel_format(style='plain')
        except:
            pass

1	# /`-
2	# _ _ _ /####`-
3	# \| \| \| \| (_) /########`-
4	# \| \|_ _ __ __ _ _ __ ___ ___ _ __ \| \|_ _ ___ /###########`-
5	# \| __\| '__/ _` \| '_ \/ __\|/ _ \ '_ \\| __\| / __\| ____ -###########/
6	# \| \|_\| \| \| (_\| \| \| \| \__ \ __/ \| \| \| \|_\| \__ \ \| \| `-#######/
7	# \__\|_\| \__,_\|_\| \|_\|___/\___\|_\| \|_\|\__\|_\|___/ \|____\| `- # /
8	#
9	# Copyright (c) 2018 transentis labs GmbH
10	# MIT License
11
12
13
14	import statistics	1✔
15	import pandas as pd	1✔
16
17
18	class visualizer():	1✔
19	"""
20	Class for building plots from dataframes. Includes capabilities to produce time series data.
21	Can also only modify dataframes to generate time series and return the modified dataframe
22	"""
23
24	def __init__(self,config=None):	1✔
25	self.config = config	1✔
26
27	def plot(self, df, return_df, visualize_from_period, visualize_to_period, stacked, kind, title, alpha, x_label,	1✔
28	y_label, start_date="1/1/2018", freq="D", series_names={},format="plot"):
29	"""
30	Plot method. Creates plots from dataframes
31	:param df: DataFrame input
32	:param return_df: Flag. If true return a dataFrame and do not plot (default: False)
33	:param visualize_from_period: visualize from a specific t (default: 0)
34	:param visualize_to_period: visualize until a specific t (default: model's stoptime)
35	:param stacked: If True, use stacked series (default: False)
36	:param kind: 'area' or 'line' plot (ldefault: area)
37	:param title: Title of plot
38	:param alpha: Alpha of series (default: see config!)
39	:param x_label: x_label of plot
40	:param y_label: y_label of plot
41	:param start_date: Start date for time series
42	:param freq: Frequency setting for time series
43	:param series_names: series renaming patterns
44	:return: depends on format flag: either just a plot, which formally returns nothing, or Matplotlib axes, or a dataframe
45	"""
46
47	if not kind:	1✔
48	kind=self.config.configuration["kind"]	1✔
49
50	if not stacked:	1✔
51	stacked = self.config.configuration["stacked"]	1✔
52
53	if not alpha:	1✔
54	alpha = self.config.configuration["alpha"]	1✔
55
56	if not start_date == "":	1✔
57	df.index = pd.date_range(start_date, periods=len(df), freq=freq)	1✔
58
59	series_names_keys = series_names.keys()	1✔
60
61	if len(series_names) > 0:	1✔
62
63
64	new_columns = {}	1✔
65	for column in df.columns:	1✔
66	for series_names_key in series_names_keys:	1✔
67	if series_names_key in column:	1✔
68	new_column = column.replace(series_names_key, series_names[series_names_key])	1✔
69	new_columns[column] = new_column	1✔
70
71	df.rename(columns=new_columns, inplace=True)	1✔
72
73	## If user did not set return_df=True, plot the simulation results (default behavior)
74	if not (return_df or format=="df"):	1✔
75
76	### Get the plot object
77	if visualize_to_period == 0:	1✔
78
79	ax = df.iloc[visualize_from_period:].plot(kind=kind, stacked=stacked,	1✔
80	figsize=self.config.configuration["figsize"],
81	title=title,
82	alpha=alpha, color=self.config.configuration["colors"],
83	lw=self.config.configuration["linewidth"])
84
85	elif visualize_from_period == visualize_to_period:	1✔
86	print("[INFO] No data to plot for period t={} to t={}".format(str(visualize_from_period),	1✔
87	str(visualize_to_period)))
88	return None	1✔
89
90	else:
91	if visualize_to_period + 1 > len(df):	1✔
92	visualize_to_period = len(df)	1✔
93
94	ax = df.iloc[visualize_from_period:visualize_to_period].plot(kind=kind, stacked=stacked,	1✔
95	figsize=self.config.configuration["figsize"],
96	title=title,
97	alpha=alpha,
98	color=self.config.configuration["colors"],
99	lw=self.config.configuration["linewidth"])
100	### Set axes labels and set the formats
101	if (len(x_label) > 0):	1✔
102	ax.set_xlabel(x_label)	1✔
103
104	# Set the y-axis label
105	if (len(y_label) > 0):	1✔
106	ax.set_ylabel(y_label)	1✔
107
108	for ymaj in ax.yaxis.get_majorticklocs():	1✔
109	ax.axhline(y=ymaj, ls='-', alpha=0.05, color=(34.1 / 100, 32.9 / 100, 34.1 / 100))	1✔
110
111	self.update_plot_formats(ax)	1✔
112
113	if format=="axes":	1✔
114	return ax	×
115	else:
116	return	1✔
117
118	### If user wanted a dataframe instead, here it is!
119	elif return_df or format=="df":	1✔
120	if visualize_to_period == 0:	1✔
121	return df.iloc[visualize_from_period:]	1✔
122	elif visualize_from_period == visualize_to_period:	1✔
123	print("[INFO] No data for period t={} to t={}".format(str(visualize_from_period + 1),	1✔
124	str(visualize_to_period + 1)))
125	return None	1✔
126	else:
127	return df.iloc[visualize_from_period:visualize_to_period]	1✔
128
129
130
131	def update_plot_formats(self, ax):	1✔
132	"""
133	Configure the plot formats for the labels. Generates the formatting for y labels
134	:param ax:
135	:return:
136	"""
137	ylabels_mean = statistics.mean(ax.get_yticks())	1✔
138
139
140	# Override the format based on the mean values
141
142	import matplotlib.ticker as ticker	1✔
143
144	def label_format(x,pos):	1✔
145	if ylabels_mean <= 2.0 and ylabels_mean >= -2.0:	1✔
146	label = format(x, ',.2f')	1✔
147
148	elif ylabels_mean <= 10.0 and ylabels_mean >= -10.0:	1✔
149	label = format(x, ',.1f')	1✔
150
151	else:
152	label = format(x, ',.0f')	1✔
153
154	return label	1✔
155
156
157	ax.yaxis.set_major_formatter(ticker.FuncFormatter(func=label_format))	1✔
158
159	## Quick fix for incompatibility of scientific notation and ticklabels
160	try:	1✔
161	ax.ticklabel_format(style='plain')	1✔
162	except:	1✔
163	pass	1✔

transentis / bptk_py / 18773402293

Source File Press 'n' to go to next uncovered line, 'b' for previous

Source File
Press 'n' to go to next uncovered line, 'b' for previous