Replace np.NaN with np.nan (#8)

* wip

* Linter issues

Author: gedeck

setup.py

@@ -49,6 +49,7 @@ def getVersion():
     install_requires=[
         'pyDOE3',
         'diversipy',
+        'pingouin',
     ],
 
     test_suite='nose.collector',

src/mistat/design/doepy/pydoe_corrected.py

@@ -538,24 +538,30 @@ def ccdesign_corrected(n, center=(4, 4), alpha="orthogonal", face="circumscribed
         H2, a = star(n, alpha="orthogonal", center=center)
 
     # Rotatable Design
-    if alpha.lower() in ("rotatable", "r"):
+    elif alpha.lower() in ("rotatable", "r"):
         H2, a = star(n, alpha="rotatable")
 
+    else:
+        raise NotImplementedError(f'Invalid value for "alpha": {alpha}')
+
     # Inscribed CCD
     if face.lower() in ("inscribed", "cci"):
         H1 = ff2n_corrected(n)
         H1 = H1 / a  # Scale down the factorial points
         H2, a = star(n)
 
     # Faced CCD
-    if face.lower() in ("faced", "ccf"):
+    elif face.lower() in ("faced", "ccf"):
         H2, a = star(n)  # Value of alpha is always 1 in Faced CCD
         H1 = ff2n_corrected(n)
 
     # Circumscribed CCD
-    if face.lower() in ("circumscribed", "ccc"):
+    elif face.lower() in ("circumscribed", "ccc"):
         H1 = ff2n_corrected(n)
 
+    else:
+        raise NotImplementedError(f'Invalid value for "face": {face}')
+
     C1 = repeat_center(n, center[0])
     C2 = repeat_center(n, center[1])
 

src/mistat/mqcc/tests/test_statistics.py

@@ -15,7 +15,7 @@
 
 
 class Test_mqccStatistics(unittest.TestCase):
-    array_1 = np.array([[1, 2, 3], [1, 2, np.NaN]])
+    array_1 = np.array([[1, 2, 3], [1, 2, np.nan]])
     array_2 = np.array([[1, 2, 3], [1, 2, 7]])
 
     def setUp(self):

src/mistat/qcc/processCapability.py

@@ -50,10 +50,10 @@ def __init__(self, qcc, spec_limits, std_dev=None, target=None, nsigmas=None,
         self.confidence_level = confidence_level
         self.calcPCIndices()
 
-        self.exp_LSL = np.NaN
+        self.exp_LSL = np.nan
         if not np.isnan(lsl):
             self.exp_LSL = stats.norm.cdf((lsl - self.center) / self.std_dev) * 100
-        self.exp_USL = np.NaN
+        self.exp_USL = np.nan
         if not np.isnan(usl):
             self.exp_USL = (1 - stats.norm.cdf((usl - self.center) / self.std_dev)) * 100
         self.obs_LSL = np.nanmean(data < lsl) * 100

src/mistat/qcc/qualityControlChart.py

@@ -102,7 +102,7 @@ def plot(self, title=None, ax=None):
             ucl = [[], []]
             last = None
             for xleft, xright, lower, upper in zip(df.x.values - delta[:-1], df.x.values + delta[1:],
-                                           self.limits.LCL, self.limits.UCL):
+                                                   self.limits.LCL, self.limits.UCL):
                 if last:
                     lcl[0].extend([xleft, xleft])
                     lcl[1].extend([last, lower])
@@ -182,7 +182,7 @@ def qcc_groups(data, groups):
     max_size = max(grouped.size())
 
     result = np.empty((len(grouped), max_size))
-    result[:] = np.NaN
+    result[:] = np.nan
     for idx, group in grouped:
         result[idx - 1, :len(group.values)] = group.values
     return result

src/mistat/qcc/rules.py

@@ -35,8 +35,8 @@ def beyondLimits(qcc, limits=None, **kwargs):
     if len(limits['UCL']) == 1:
         return {'beyondLimits': {'UCL': np.nonzero(statistics > limits['UCL'][0])[0],
                                  'LCL': np.nonzero(statistics < limits['LCL'][0])[0]}}
-    return {'beyondLimits': {'UCL': np.nonzero(statistics.ravel() > limits['UCL'].ravel())[0],
-                             'LCL': np.nonzero(statistics.ravel() < limits['LCL'].ravel())[0]}}
+    return {'beyondLimits': {'UCL': np.nonzero(statistics > limits['UCL'].to_numpy())[0],
+                             'LCL': np.nonzero(statistics < limits['LCL'].to_numpy())[0]}}
 
 
 def violatingRuns(qcc, run_length=7, **kwargs):

src/mistat/qcc/statistics.py

@@ -90,12 +90,12 @@ def get(cls, name, default):
 # exp.R.unscaled a vector specifying, for each sample size, the expected value of the relative range
 # (i.e. R/σ) for a normal distribution. This appears as d2 on most tables containing factors for
 # the construction of control charts.
-_exp_R_unscaled = [np.NaN, np.NaN, 1.128, 1.693, 2.059, 2.326, 2.534, 2.704, 2.847, 2.970, 3.078, 3.173,
+_exp_R_unscaled = [np.nan, np.nan, 1.128, 1.693, 2.059, 2.326, 2.534, 2.704, 2.847, 2.970, 3.078, 3.173,
                    3.258, 3.336, 3.407, 3.472, 3.532, 3.588, 3.640, 3.689, 3.735, 3.778, 3.819, 3.858, 3.895, 3.931]
 # se.R.unscaled a vector specifying, for each sample size, the standard error of the relative range
 # (i.e. R/σ) for a normal distribution. This appears as d3 on most tables containing factors for
 # the construction of control charts.
-_se_R_unscaled = [np.NaN, np.NaN, 0.8525033, 0.8883697, 0.8798108, 0.8640855, 0.8480442, 0.8332108, 0.8198378,
+_se_R_unscaled = [np.nan, np.nan, 0.8525033, 0.8883697, 0.8798108, 0.8640855, 0.8480442, 0.8332108, 0.8198378,
                   0.8078413, 0.7970584, 0.7873230, 0.7784873, 0.7704257, 0.7630330, 0.7562217, 0.7499188, 0.7440627,
                   0.7386021, 0.7334929, 0.7286980, 0.7241851, 0.7199267, 0.7158987, 0.7120802, 0.7084528, 0.7050004,
                   0.7017086, 0.6985648, 0.6955576, 0.6926770, 0.6899137, 0.6872596, 0.6847074, 0.6822502, 0.6798821,

src/mistat/qcc/tests/test_qualityControlChart.py

@@ -27,9 +27,9 @@ def test_qcc_groups(num_regression):
 
 def test_qcc_groups_missing_values():
     result = qcc_groups([1, 2, 2, 3, 3, 3, 4, 4, 4, 4], [1, 2, 2, 3, 3, 3, 4, 4, 4, 4])
-    expect = np.array([[1.0, np.NaN, np.NaN, np.NaN],
-                       [2.0, 2.0, np.NaN, np.NaN],
-                       [3.0, 3.0, 3.0, np.NaN],
+    expect = np.array([[1.0, np.nan, np.nan, np.nan],
+                       [2.0, 2.0, np.nan, np.nan],
+                       [3.0, 3.0, 3.0, np.nan],
                        [4.0, 4.0, 4.0, 4.0]])
     np.testing.assert_array_equal(result, expect)
 

src/mistat/qcc/tests/test_statistics.py

@@ -16,7 +16,7 @@
 
 
 class Test_qccStatistics(unittest.TestCase):
-    array_1 = np.array([[1, 2, 3], [1, 2, np.NaN]])
+    array_1 = np.array([[1, 2, 3], [1, 2, np.nan]])
     array_2 = np.array([[1, 2, 3], [1, 2, 7]])
 
     def setUp(self):

src/mistat/runsTest.py

@@ -42,6 +42,8 @@ def runsTest(sequence, cutoff=None, alternative=None, verbose=False):
         pval = stats.norm.cdf(Z)
     elif alternative == 'greater':
         pval = stats.norm.cdf(-Z)
+    else:
+        raise NotImplementedError('alternative not implemented')
 
     Result = namedtuple('Result', 'statistic,pval,method,alternative')
     result = Result(Z, pval, 'Runs Test', alternative)

src/mistat/version.py

@@ -2,6 +2,6 @@
 Modern Statistics: A Computer Based Approach with Python
 Industrial Statistics: A Computer Based Approach with Python
 
-(c) 2022 Ron Kenett, Shelemyahu Zacks, Peter Gedeck
+(c) 2022-2024 Ron Kenett, Shelemyahu Zacks, Peter Gedeck
 '''
-__version__ = '0.1.14'
+__version__ = '0.1.16'

src/pylintrc

@@ -69,6 +69,7 @@ disable=missing-function-docstring,
         import-outside-toplevel,
         too-many-locals,
         too-many-statements,
+        too-many-positional-arguments,
         logging-fstring-interpolation,
         fixme,
         too-many-branches,