knowyourdata.kyd.KYD.get_basic_stats() - Code Metrics - Inspection of "Added support for full non-finite arrays as well a..." - mubdi/knowyourdata - Measure and Improve Code Quality continuously with Scrutinizer

Completed

Push — master ( 15cfe4...5ce4a0 )

by Mubdi

created 2018-03-13 14:41 UTC

knowyourdata.kyd.KYD.get_basic_stats() B

↳ Parent: knowyourdata.kyd

Complexity

Conditions

Size

Total Lines	25
Code Lines	21

Duplication

Lines	0
Ratio	0 %

Importance

Changes

Metric	Value
cc	2
eloc	21
nop	1
dl	0
loc	25
rs	8.8571
c	0
b	0
f	0

"""
KnowYourData
============

A rapid and lightweight module to describe the statistics and structure of
data arrays for interactive use.

The most simple use case to display data is if you have a numpy array 'x':

    >>> from knowyourdata import kyd
    >>> kyd(x)

"""

import sys
import numpy as np


class KYD(object):

    """The Central Class for KYD"""

    # Variable for Data Vector
    data = None

    # Initial Flags
    f_allfinite = False
    f_allnonfinite = False
    f_hasnan = False
    f_hasinf = False

    # Display Settings
    col_width = 10
    precision = 4

    def check_finite(self):
        """Checking to see if all elements are finite and setting flags"""
        if np.all(np.isfinite(self.data)):
            self.filt_data = self.data
            self.f_allfinite = True
        else:
            finite_inds = np.where(np.isfinite(self.data))

            self.filt_data = self.data[finite_inds]

            if self.filt_data.size == 0:
                self.f_allnonfinite = True

            if np.any(np.isnan(self.data)):
                self.f_hasnan = True
            if np.any(np.isinf(self.data)):
                self.f_hasinf = True

    def check_struct(self):
        """Determining the Structure of the Numpy Array"""
        self.dtype = self.data.dtype
        self.ndim = self.data.ndim
        self.shape = self.data.shape
        self.size = self.data.size
        self.memsize = sys.getsizeof(self.data)
        self.human_memsize = sizeof_fmt(self.memsize)

    def get_basic_stats(self):
        """Get basic statistics about array"""

        if self.f_allnonfinite:
            self.min = self.max = self.range = np.nan
            self.mean = self.std = self.median = np.nan
            self.firstquartile = self.thirdquartile = np.nan
            self.cl_68 = self.cl_95 = self.cl_99 = np.array([np.nan, np.nan])

            return

        self.min = np.float_(np.min(self.filt_data))
        self.max = np.float_(np.max(self.filt_data))
        self.range = self.max - self.min
        self.mean = np.mean(self.filt_data)
        self.std = np.std(self.filt_data)
        self.median = np.float_(np.median(self.filt_data))
        self.firstquartile = np.float_(np.percentile(self.filt_data, 25))
        self.thirdquartile = np.float_(np.percentile(self.filt_data, 75))
        self.cl_99 = np.float_(
            np.percentile(self.filt_data, np.array([0.5, 99.5])))
        self.cl_95 = np.float_(
            np.percentile(self.filt_data, np.array([2.5, 97.5])))
        self.cl_68 = np.float_(
            np.percentile(self.filt_data, np.array([16.0, 84.0])))

    def display_basic_stats(self):
        """Display basic statistics of array"""
        pstr_list = []

        # Heading for Section

        pstr_struct_header1 = '\033[1m' + "Basic Statistics  " + '\033[0m'
        pstr_struct_header2 = ''

        pstr_list.append(pstr_struct_header1)
        pstr_list.append(pstr_struct_header2)

        # Mean and Standard Deviation

        pstr_meanstdhead = (
            "{0:^15}"
            "{1:^15}"
        ).format("Mean", "Std Dev")
        pstr_meanstdhead = (
            "{0:^{self.col_width}}"
        ).format(pstr_meanstdhead, self=self)
        pstr_list.append(pstr_meanstdhead)

        pstr_meanstdstat = (
            "{self.mean:^15.{self.precision}}"
            "{self.std:^15.{self.precision}}"
        ).format(self=self)
        pstr_meanstdstat = (
            "{0:^{self.col_width}}"
        ).format(pstr_meanstdstat, self=self)
        pstr_list.append(pstr_meanstdstat)

        pstr_list.append("")

        # Three point statistics

        pstr_3pthead = (
            "{0:^10}"
            "{1:^10}"
            "{2:^10}"
            "{3:^10}"
            "{4:^10}"
        ).format('Min', '1Q', 'Median', '3Q', 'Max')
        pstr_3pthead = (
            "{0:^{self.col_width}}"
        ).format(pstr_3pthead, self=self)
        pstr_list.append(pstr_3pthead)

        pstr_3ptstat = (
            "{self.min:^10.{self.precision}}"
            "{self.firstquartile:^10.{self.precision}}"
            "{self.median:^10.{self.precision}}"
            "{self.thirdquartile:^10.{self.precision}}"
            "{self.max:^10.{self.precision}}"
        ).format(self=self)
        pstr_3ptstat = (
            "{0:^{self.col_width}}"
        ).format(pstr_3ptstat, self=self)
        pstr_list.append(pstr_3ptstat)

        pstr_list.append("")

        # Confidence Levels

        pstr_clhead = (
            "{0:^10}"
            "{1:^10}"
            "{2:^10}"
            "{3:^10}"
            "{4:^10}"
            "{5:^10}"
        ).format('-99 CL', '-95 CL', '-68 CL', '+68 CL', '+95 CL', '+99 CL')
        pstr_clhead = (
            "{0:^{self.col_width}}"
        ).format(pstr_clhead, self=self)
        pstr_list.append(pstr_clhead)

        pstr_clstat = (
            "{self.cl_99[0]:^10.{self.precision}}"
            "{self.cl_95[0]:^10.{self.precision}}"
            "{self.cl_68[0]:^10.{self.precision}}"
            "{self.cl_68[1]:^10.{self.precision}}"
            "{self.cl_95[1]:^10.{self.precision}}"
            "{self.cl_99[1]:^10.{self.precision}}"
        ).format(self=self)
        pstr_clstat = (
            "{0:^{self.col_width}}"
        ).format(pstr_clstat, self=self)
        pstr_list.append(pstr_clstat)

        return pstr_list

    def display_struct(self):
        """Display information about array structure"""

        pstr_list = []

        # pstr_struct_header0 = "................."
        pstr_struct_header1 = '\033[1m' + "Array Structure  " + '\033[0m'
        pstr_struct_header2 = "                 "

        # pstr_list.append(pstr_struct_header0)
        pstr_list.append(pstr_struct_header1)
        pstr_list.append(pstr_struct_header2)

        pstr_n_dim = (
            "Number of Dimensions:\t"
            "{self.ndim}").format(
                self=self)
        pstr_list.append(pstr_n_dim)

        pstr_shape = (
            "Shape of Dimensions:\t"
            "{self.shape}").format(
                self=self)
        pstr_list.append(pstr_shape)

        pstr_dtype = (
            "Array Data Type:\t"
            "{self.dtype}").format(
                self=self)
        pstr_list.append(pstr_dtype)

        pstr_memsize = (
            "Memory Size:\t\t"
            "{self.human_memsize}").format(
                self=self)
        pstr_list.append(pstr_memsize)

        return pstr_list

    def display(self, short=False):
        """Displaying all relevant statistics"""

        if short:
            pass

        print("")
        pstr_basic = self.display_basic_stats()
        pstr_struct = self.display_struct()
        n_basic = len(pstr_basic)
        n_struct = len(pstr_struct)

        l_colwidth = max([len(x) for x in pstr_basic]) + 1

        r_colwidth = max([len(x) for x in pstr_struct]) + 2

        # new_colwidth = self.col_width + 20

        # Finding the longest string
        len_list = max([n_basic, n_struct])

        for i in range(len_list):
            tmp_str = '| '
            if i < n_basic:
                tmp_str += (pstr_basic[i].ljust(l_colwidth))
            else:
                tmp_str += ''.ljust(l_colwidth)
            tmp_str += '\t| '

            if i < n_struct:
                tmp_str += (pstr_struct[i].expandtabs().ljust(r_colwidth))
            else:
                tmp_str += ''.ljust(r_colwidth)
            tmp_str += '\t|'

            print(tmp_str)

        print("")

    def clear_memory(self):
        """Ensuring the Numpy Array does not exist in memory"""
        del self.data
        del self.filt_data

    def __init__(self, data):
        super(KYD, self).__init__()

        # Ensuring that the array is a numpy array
        if not isinstance(data, np.ndarray):
            data = np.array(data)

        self.data = data

        self.check_finite()
        self.check_struct()
        self.get_basic_stats()
        self.clear_memory()


def sizeof_fmt(num, suffix='B'):
    """Return human readable version of in-memory size.
    Code from Fred Cirera from Stack Overflow:
    https://stackoverflow.com/questions/1094841/reusable-library-to-get-human-readable-version-of-file-size
    """
    for unit in ['', 'Ki', 'Mi', 'Gi', 'Ti', 'Pi', 'Ei', 'Zi']:
        if abs(num) < 1024.0:
            return "%3.1f%s%s" % (num, unit, suffix)
        num /= 1024.0
    return "%.1f%s%s" % (num, 'Yi', suffix)


def kyd(data, full_statistics=False):
    """Print statistics of any numpy array

    data -- Numpy Array of Data

    Keyword arguments:
    full_statistics -- printing all detailed statistics of the sources
    (Currently Not Implemented)

    """

    data_kyd = KYD(data)
    if full_statistics:
        data_kyd.display()
    else:
        data_kyd.display(short=True)

    return data_kyd


1			"""
2			KnowYourData
3			============
4
5			A rapid and lightweight module to describe the statistics and structure of
6			data arrays for interactive use.
7
8			The most simple use case to display data is if you have a numpy array 'x':
9
10			>>> from knowyourdata import kyd
11			>>> kyd(x)
12
13			"""
14
15			import sys
16			import numpy as np
17
18
19			class KYD(object):
			0 ignored issues – show Unused Code introduced 2018-03-12 18:01 UTC by Report Bug Copy Issue Report The variable `__class__` seems to be unused. Loading history... best-practice introduced 2018-03-13 14:45 UTC by Report Bug Copy Issue Report Too many instance attributes (23/7) Loading history...
20			"""The Central Class for KYD"""
21
22			# Variable for Data Vector
23			data = None
24
25			# Initial Flags
26			f_allfinite = False
27			f_allnonfinite = False
28			f_hasnan = False
29			f_hasinf = False
30
31			# Display Settings
32			col_width = 10
33			precision = 4
34
35			def check_finite(self):
36			"""Checking to see if all elements are finite and setting flags"""
37			if np.all(np.isfinite(self.data)):
38			self.filt_data = self.data
39			self.f_allfinite = True
40			else:
41			finite_inds = np.where(np.isfinite(self.data))
42
43			self.filt_data = self.data[finite_inds]
44
45			if self.filt_data.size == 0:
46			self.f_allnonfinite = True
47
48			if np.any(np.isnan(self.data)):
49			self.f_hasnan = True
50			if np.any(np.isinf(self.data)):
51			self.f_hasinf = True
52
53			def check_struct(self):
54			"""Determining the Structure of the Numpy Array"""
55			self.dtype = self.data.dtype
56			self.ndim = self.data.ndim
57			self.shape = self.data.shape
58			self.size = self.data.size
59			self.memsize = sys.getsizeof(self.data)
60			self.human_memsize = sizeof_fmt(self.memsize)
61
62			def get_basic_stats(self):
63			"""Get basic statistics about array"""
64
65			if self.f_allnonfinite:
66			self.min = self.max = self.range = np.nan
67			self.mean = self.std = self.median = np.nan
68			self.firstquartile = self.thirdquartile = np.nan
69			self.cl_68 = self.cl_95 = self.cl_99 = np.array([np.nan, np.nan])
70
71			return
72
73			self.min = np.float_(np.min(self.filt_data))
74			self.max = np.float_(np.max(self.filt_data))
75			self.range = self.max - self.min
76			self.mean = np.mean(self.filt_data)
77			self.std = np.std(self.filt_data)
78			self.median = np.float_(np.median(self.filt_data))
79			self.firstquartile = np.float_(np.percentile(self.filt_data, 25))
80			self.thirdquartile = np.float_(np.percentile(self.filt_data, 75))
81			self.cl_99 = np.float_(
82			np.percentile(self.filt_data, np.array([0.5, 99.5])))
83			self.cl_95 = np.float_(
84			np.percentile(self.filt_data, np.array([2.5, 97.5])))
85			self.cl_68 = np.float_(
86			np.percentile(self.filt_data, np.array([16.0, 84.0])))
87
88			def display_basic_stats(self):
89			"""Display basic statistics of array"""
90			pstr_list = []
91
92			# Heading for Section
93
94			pstr_struct_header1 = '\033[1m' + "Basic Statistics " + '\033[0m'
95			pstr_struct_header2 = ''
96
97			pstr_list.append(pstr_struct_header1)
98			pstr_list.append(pstr_struct_header2)
99
100			# Mean and Standard Deviation
101
102			pstr_meanstdhead = (
103			"{0:^15}"
104			"{1:^15}"
105			).format("Mean", "Std Dev")
106			pstr_meanstdhead = (
107			"{0:^{self.col_width}}"
108			).format(pstr_meanstdhead, self=self)
109			pstr_list.append(pstr_meanstdhead)
110
111			pstr_meanstdstat = (
112			"{self.mean:^15.{self.precision}}"
113			"{self.std:^15.{self.precision}}"
114			).format(self=self)
115			pstr_meanstdstat = (
116			"{0:^{self.col_width}}"
117			).format(pstr_meanstdstat, self=self)
118			pstr_list.append(pstr_meanstdstat)
119
120			pstr_list.append("")
121
122			# Three point statistics
123
124			pstr_3pthead = (
125			"{0:^10}"
126			"{1:^10}"
127			"{2:^10}"
128			"{3:^10}"
129			"{4:^10}"
130			).format('Min', '1Q', 'Median', '3Q', 'Max')
131			pstr_3pthead = (
132			"{0:^{self.col_width}}"
133			).format(pstr_3pthead, self=self)
134			pstr_list.append(pstr_3pthead)
135
136			pstr_3ptstat = (
137			"{self.min:^10.{self.precision}}"
138			"{self.firstquartile:^10.{self.precision}}"
139			"{self.median:^10.{self.precision}}"
140			"{self.thirdquartile:^10.{self.precision}}"
141			"{self.max:^10.{self.precision}}"
142			).format(self=self)
143			pstr_3ptstat = (
144			"{0:^{self.col_width}}"
145			).format(pstr_3ptstat, self=self)
146			pstr_list.append(pstr_3ptstat)
147
148			pstr_list.append("")
149
150			# Confidence Levels
151
152			pstr_clhead = (
153			"{0:^10}"
154			"{1:^10}"
155			"{2:^10}"
156			"{3:^10}"
157			"{4:^10}"
158			"{5:^10}"
159			).format('-99 CL', '-95 CL', '-68 CL', '+68 CL', '+95 CL', '+99 CL')
160			pstr_clhead = (
161			"{0:^{self.col_width}}"
162			).format(pstr_clhead, self=self)
163			pstr_list.append(pstr_clhead)
164
165			pstr_clstat = (
166			"{self.cl_99[0]:^10.{self.precision}}"
167			"{self.cl_95[0]:^10.{self.precision}}"
168			"{self.cl_68[0]:^10.{self.precision}}"
169			"{self.cl_68[1]:^10.{self.precision}}"
170			"{self.cl_95[1]:^10.{self.precision}}"
171			"{self.cl_99[1]:^10.{self.precision}}"
172			).format(self=self)
173			pstr_clstat = (
174			"{0:^{self.col_width}}"
175			).format(pstr_clstat, self=self)
176			pstr_list.append(pstr_clstat)
177
178			return pstr_list
179
180			def display_struct(self):
181			"""Display information about array structure"""
182
183			pstr_list = []
184
185			# pstr_struct_header0 = "................."
186			pstr_struct_header1 = '\033[1m' + "Array Structure " + '\033[0m'
187			pstr_struct_header2 = " "
188
189			# pstr_list.append(pstr_struct_header0)
190			pstr_list.append(pstr_struct_header1)
191			pstr_list.append(pstr_struct_header2)
192
193			pstr_n_dim = (
194			"Number of Dimensions:\t"
195			"{self.ndim}").format(
196			self=self)
197			pstr_list.append(pstr_n_dim)
198
199			pstr_shape = (
200			"Shape of Dimensions:\t"
201			"{self.shape}").format(
202			self=self)
203			pstr_list.append(pstr_shape)
204
205			pstr_dtype = (
206			"Array Data Type:\t"
207			"{self.dtype}").format(
208			self=self)
209			pstr_list.append(pstr_dtype)
210
211			pstr_memsize = (
212			"Memory Size:\t\t"
213			"{self.human_memsize}").format(
214			self=self)
215			pstr_list.append(pstr_memsize)
216
217			return pstr_list
218
219			def display(self, short=False):
220			"""Displaying all relevant statistics"""
221
222			if short:
223			pass
224
225			print("")
226			pstr_basic = self.display_basic_stats()
227			pstr_struct = self.display_struct()
228			n_basic = len(pstr_basic)
229			n_struct = len(pstr_struct)
230
231			l_colwidth = max([len(x) for x in pstr_basic]) + 1
232
233			r_colwidth = max([len(x) for x in pstr_struct]) + 2
234
235			# new_colwidth = self.col_width + 20
236
237			# Finding the longest string
238			len_list = max([n_basic, n_struct])
239
240			for i in range(len_list):
241			tmp_str = '\| '
242			if i < n_basic:
243			tmp_str += (pstr_basic[i].ljust(l_colwidth))
244			else:
245			tmp_str += ''.ljust(l_colwidth)
246			tmp_str += '\t\| '
247
248			if i < n_struct:
249			tmp_str += (pstr_struct[i].expandtabs().ljust(r_colwidth))
250			else:
251			tmp_str += ''.ljust(r_colwidth)
252			tmp_str += '\t\|'
253
254			print(tmp_str)
255
256			print("")
257
258			def clear_memory(self):
259			"""Ensuring the Numpy Array does not exist in memory"""
260			del self.data
261			del self.filt_data
262
263			def __init__(self, data):
264			super(KYD, self).__init__()
265
266			# Ensuring that the array is a numpy array
267			if not isinstance(data, np.ndarray):
268			data = np.array(data)
269
270			self.data = data
271
272			self.check_finite()
273			self.check_struct()
274			self.get_basic_stats()
275			self.clear_memory()
276
277
278			def sizeof_fmt(num, suffix='B'):
279			"""Return human readable version of in-memory size.
280			Code from Fred Cirera from Stack Overflow:
281			https://stackoverflow.com/questions/1094841/reusable-library-to-get-human-readable-version-of-file-size
282			"""
283			for unit in ['', 'Ki', 'Mi', 'Gi', 'Ti', 'Pi', 'Ei', 'Zi']:
284			if abs(num) < 1024.0:
285			return "%3.1f%s%s" % (num, unit, suffix)
286			num /= 1024.0
287			return "%.1f%s%s" % (num, 'Yi', suffix)
288
289
290			def kyd(data, full_statistics=False):
291			"""Print statistics of any numpy array
292
293			data -- Numpy Array of Data
294
295			Keyword arguments:
296			full_statistics -- printing all detailed statistics of the sources
297			(Currently Not Implemented)
298
299			"""
300
301			data_kyd = KYD(data)
302			if full_statistics:
303			data_kyd.display()
304			else:
305			data_kyd.display(short=True)
306
307			return data_kyd
308

mubdi / knowyourdata

Push — master ( 15cfe4...5ce4a0 )

knowyourdata.kyd.KYD.get_basic_stats() B

Complexity

Size

Duplication

Importance

Duplication Side-by-Side

Filter issues like