Merge pull request #171 from arokem/mpl_iterable

Address warnings

Author: arokem

doc/examples/resting_state_fmri.py

@@ -31,7 +31,6 @@
 #Import from other libraries:
 import numpy as np
 import matplotlib.pyplot as plt
-from matplotlib.mlab import csv2rec
 
 import nitime
 #Import the time-series objects:
@@ -49,14 +48,12 @@
 
 """
 
-We use csv2rec to read the data in from file to a recarray:
+We use Numpy to read the data in from file to a recarray:
 
 """
 
 data_path = os.path.join(nitime.__path__[0], 'data')
 
-data_rec = csv2rec(os.path.join(data_path, 'fmri_timeseries.csv'))
-
 data_rec = np.genfromtxt(os.path.join(data_path, 'fmri_timeseries.csv'),
                          names=True, delimiter=',')
 

nitime/algorithms/cohere.py

@@ -975,7 +975,7 @@ def cache_fft(time_series, ij, lb=0, ub=None,
 
     n_freqs = ub_idx - lb_idx
     # Make the window:
-    if mlab.cbook.iterable(window):
+    if np.iterable(window):
         assert(len(window) == NFFT)
         window_vals = window
     else:

nitime/algorithms/event_related.py

@@ -55,9 +55,9 @@ def fir(timeseries, design):
     and Unbiased Approach. Human Brain Mapping, 11:249-260
 
     """
-    X = np.matrix(design)
-    y = np.matrix(timeseries)
-    h = np.array(linalg.pinv(X.T * X) * X.T * y.T)
+    X = np.array(design)
+    y = np.array(timeseries)
+    h = np.array(linalg.pinv(X.T @ X) @ X.T @ y.T)
     return h
 
 

nitime/algorithms/spectral.py

@@ -105,7 +105,7 @@ def get_spectra(time_series, method=None):
         Fs = method.get('Fs', 2 * np.pi)
         detrend = method.get('detrend', mlab.detrend_none)
         window = method.get('window', mlab.window_hanning)
-        n_overlap = method.get('n_overlap', int(np.ceil(NFFT / 2)))
+        n_overlap = method.get('n_overlap', int(np.ceil(NFFT //  2)))
 
         # The length of the spectrum depends on how many sides are taken, which
         # depends on whether or not this is a complex object:
@@ -587,7 +587,7 @@ def multi_taper_psd(
     sdf_est /= Fs
 
     if sides == 'onesided':
-        freqs = np.linspace(0, Fs / 2, NFFT / 2 + 1)
+        freqs = np.linspace(0, Fs / 2, NFFT //  2 + 1)
     else:
         freqs = np.linspace(0, Fs, NFFT, endpoint=False)
 
@@ -726,7 +726,7 @@ def multi_taper_csd(s, Fs=2 * np.pi, NW=None, BW=None, low_bias=True,
     csdfs /= Fs
 
     if sides == 'onesided':
-        freqs = np.linspace(0, Fs / 2, NFFT / 2 + 1)
+        freqs = np.linspace(0, Fs / 2, NFFT //  2 + 1)
     else:
         freqs = np.linspace(0, Fs, NFFT, endpoint=False)
 

nitime/algorithms/tests/test_autoregressive.py

@@ -1,6 +1,5 @@
 import numpy as np
 import numpy.testing as npt
-import numpy.testing.decorators as dec
 
 import nitime.algorithms as tsa
 import nitime.utils as utils
@@ -87,7 +86,6 @@ def test_AR_LD():
     npt.assert_almost_equal(avg_pwr, avg_pwr_est, decimal=0)
 
 
-@dec.slow
 def test_MAR_est_LWR():
     """
 

nitime/algorithms/tests/test_spectral.py

@@ -7,7 +7,6 @@
 import scipy
 from scipy import fftpack
 import numpy.testing as npt
-import numpy.testing.decorators as dec
 import pytest
 
 import nitime.algorithms as tsa
@@ -320,7 +319,6 @@ def test_mtm_cross_spectrum():
         tsa.mtm_cross_spectrum(tspectra, np.r_[tspectra, tspectra], (w, w))
 
 
-@dec.slow
 def test_multi_taper_psd_csd():
     """
 

nitime/analysis/base.py

@@ -1,5 +1,5 @@
 
-from inspect import getargspec
+from inspect import getfullargspec
 
 from nitime import descriptors as desc
 
@@ -16,7 +16,7 @@ class BaseAnalyzer(desc.ResetMixin):
 
     @desc.setattr_on_read
     def parameterlist(self):
-        plist = getargspec(self.__init__).args
+        plist = getfullargspec(self.__init__).args
         plist.remove('self')
         plist.remove('input')
         return plist

nitime/analysis/snr.py

@@ -90,7 +90,7 @@ def __init__(self, input=None, bandwidth=None, adaptive=False,
     @desc.setattr_on_read
     def mt_frequencies(self):
         return np.linspace(0, self.input.sampling_rate / 2,
-                           self.input.data.shape[-1] / 2 + 1)
+                           self.input.data.shape[-1] // 2 + 1)
 
     @desc.setattr_on_read
     def mt_signal_psd(self):

nitime/tests/test_algorithms.py

@@ -2,7 +2,6 @@
 
 import numpy as np
 import numpy.testing as npt
-import numpy.testing.decorators as dec
 
 from scipy.signal import signaltools
 from scipy import fftpack
@@ -161,7 +160,7 @@ def test_psd_matlab():
 
     npt.assert_almost_equal(fxx_mlab, fxx_matlab, decimal=5)
 
-@dec.slow
+
 def test_long_dpss_win():
     """ Test that very long dpss windows can be generated (using interpolation)"""
 

nitime/utils.py

@@ -493,7 +493,7 @@ def adaptive_weights(yk, eigvals, sides='onesided', max_iter=150):
         """)
         # we'll hope this is a correct length for L
         N = yk.shape[-1]
-        L = N / 2 + 1 if sides == 'onesided' else N
+        L = N // 2 + 1 if sides == 'onesided' else N
         return (np.multiply.outer(np.sqrt(eigvals), np.ones(L)), 2 * K)
     rt_eig = np.sqrt(eigvals)
 
@@ -1161,8 +1161,8 @@ def fftconvolve(in1, in2, mode="full", axis=None):
     """
     s1 = np.array(in1.shape)
     s2 = np.array(in2.shape)
-    complex_result = (np.issubdtype(in1.dtype, np.complex) or
-                      np.issubdtype(in2.dtype, np.complex))
+    complex_result = (np.issubdtype(in1.dtype, np.complex128) or
+                      np.issubdtype(in2.dtype, np.complex128))
 
     if axis is None:
         size = s1 + s2 - 1
@@ -1206,10 +1206,10 @@ def fftconvolve(in1, in2, mode="full", axis=None):
 # 'get' utils
 #-----------------------------------------------------------------------------
 def get_freqs(Fs, n):
-    """Returns the center frequencies of the frequency decomposotion of a time
+    """Returns the center frequencies of the frequency decomposition of a time
     series of length n, sampled at Fs Hz"""
 
-    return np.linspace(0, float(Fs) / 2, float(n) / 2 + 1)
+    return np.linspace(0, Fs / 2, int(n / 2 + 1))
 
 
 def circle_to_hz(omega, Fsamp):
@@ -2037,8 +2037,8 @@ def zscore(time_series, axis=-1):
     st = time_series.std(axis=axis)
     sl = [slice(None)] * len(time_series.shape)
     sl[axis] = np.newaxis
-    zt = time_series - et[sl]
-    zt /= st[sl]
+    zt = time_series - et[tuple(sl)]
+    zt /= st[tuple(sl)]
     return zt