Merge pull request #369 from jakirkham/python3_support

Python 3 support

nanshe/__init__.py

@@ -210,13 +210,13 @@
     # "viewer"
 ]
 
-import box
-import converter
-import io
-import imp
-import learner
-import registerer
-import syn
-import util
-# import viewer           # Must be commented as there is some segfault
-                          # coming from Volumina.
+from nanshe import box
+from nanshe import converter
+from nanshe import io
+from nanshe import imp
+from nanshe import learner
+from nanshe import registerer
+from nanshe import syn
+from nanshe import util
+# from nanshe import viewer    # Must be commented as there is some segfault
+                               # coming from Volumina.

nanshe/box/__init__.py

@@ -18,4 +18,4 @@
 __all__ = ["spams_sandbox"]
 
 
-import spams_sandbox
+from nanshe.box import spams_sandbox

nanshe/imp/__init__.py

@@ -15,7 +15,7 @@
 
 __all__ = ["filters", "registration", "renorm", "segment"]
 
-import filters
-import registration
-import renorm
-import segment
+from nanshe.imp import filters
+from nanshe.imp import registration
+from nanshe.imp import renorm
+from nanshe.imp import segment

nanshe/imp/filters/__init__.py

@@ -16,6 +16,6 @@
 
 __all__ = ["noise", "masks", "wavelet"]
 
-import masks
-import noise
-import wavelet
+from nanshe.imp.filters import masks
+from nanshe.imp.filters import noise
+from nanshe.imp.filters import wavelet

nanshe/imp/filters/noise.py

@@ -57,19 +57,19 @@ def estimate_noise(input_array, significance_threshold=3.0):
             >>> estimate_noise(numpy.eye(2), 3)
             0.5
 
-            >>> round(estimate_noise(numpy.eye(3), 3), 3)
+            >>> round(float(estimate_noise(numpy.eye(3), 3)), 3)
             0.471
 
             >>> numpy.random.seed(10)
-            >>> round(estimate_noise(numpy.random.random((2000,2000)), 1), 3)
+            >>> round(float(estimate_noise(numpy.random.random((2000,2000)), 1)), 3)
             0.167
 
             >>> numpy.random.seed(10)
-            >>> round(estimate_noise(numpy.random.random((2000,2000)), 2), 3)
+            >>> round(float(estimate_noise(numpy.random.random((2000,2000)), 2)), 3)
             0.289
 
             >>> numpy.random.seed(10)
-            >>> round(estimate_noise(numpy.random.random((2000,2000)), 3), 3)
+            >>> round(float(estimate_noise(numpy.random.random((2000,2000)), 3)), 3)
             0.289
     """
 

nanshe/imp/filters/wavelet.py

@@ -30,6 +30,7 @@
 import vigra
 
 from nanshe.io import hdf5
+from nanshe.util.iters import irange
 from nanshe.util.xnumpy import binomial_coefficients
 
 
@@ -163,7 +164,7 @@ def binomial_1D_vigra_kernel(i, n=4, border_treatment=vigra.filters.BorderTreatm
     h_kern = binomial_1D_array_kernel(i, n)
 
     # Determine the kernel center
-    h_kern_half_width = (h_kern.size - 1) / 2
+    h_kern_half_width = (h_kern.size - 1) // 2
 
     # Default kernel
     k = vigra.filters.Kernel1D()
@@ -405,7 +406,7 @@ def transform(im0,
         imCur = im0.astype(numpy.float32)
 
 
-    for i in xrange(1, scale.max() + 1):
+    for i in irange(1, scale.max() + 1):
         if include_intermediates:
             imPrev = imCur
             imOut[i] = imOut[i - 1]
@@ -415,7 +416,7 @@ def transform(im0,
 
         h_ker = binomial_1D_vigra_kernel(i)
 
-        for d in xrange(len(scale)):
+        for d in irange(len(scale)):
             if i <= scale[d]:
                 vigra.filters.convolveOneDimension(imCur, d, h_ker, out=imCur)
 

nanshe/imp/registration.py

@@ -347,7 +347,7 @@ def register_mean_offsets(frames2reg,
         space_shift_min[space_shift_min == 0] = None
         space_shift_min = tuple(space_shift_min)
         reg_frames_slice = tuple(
-            slice(_1, _2) for _1, _2 in itertools.izip(
+            slice(_1, _2) for _1, _2 in iters.izip(
                 space_shift_max, space_shift_min
             )
         )
@@ -502,7 +502,7 @@ def translate_fourier(frame_fft, shift):
             array([[  0.,   1.,   2.,   3.],
                    [  4.,   5.,   6.,   7.],
                    [  8.,   9.,  10.,  11.]])
-            >>> af = fft.fftn(a, axes=tuple(xrange(a.ndim)))
+            >>> af = fft.fftn(a, axes=tuple(iters.irange(a.ndim)))
             >>> numpy.around(af, decimals=10)
             array([[ 66. +0.j        ,  -6. +6.j        ,  -6. +0.j        ,  -6. -6.j        ],
                    [-24.+13.85640646j,   0. +0.j        ,   0. +0.j        ,   0. +0.j        ],
@@ -517,7 +517,7 @@ def translate_fourier(frame_fft, shift):
                    [ 24.-13.85640646j,   0. -0.j        ,   0. +0.j        ,   0. +0.j        ]])
 
             >>> fft.ifftn(
-            ...     atf, axes=tuple(xrange(a.ndim))
+            ...     atf, axes=tuple(iters.irange(a.ndim))
             ... ).real.round().astype(int).astype(float)
             array([[  9.,  10.,  11.,   8.],
                    [  1.,   2.,   3.,   0.],
@@ -532,7 +532,7 @@ def translate_fourier(frame_fft, shift):
 
 
             >>> fft.ifftn(
-            ...     atf, axes=tuple(xrange(1, a.ndim))
+            ...     atf, axes=tuple(iters.irange(1, a.ndim))
             ... ).real.round().astype(int).astype(float)
             array([[[  9.,  10.,  11.,   8.],
                     [  1.,   2.,   3.,   0.],

nanshe/imp/renorm.py

@@ -25,6 +25,8 @@
 # Generally useful and fast to import so done immediately.
 import numpy
 
+from nanshe.util import iters
+
 # Need in order to have logging information no matter what.
 from nanshe.util import prof
 
@@ -225,7 +227,7 @@ def renormalized_images(input_array, ord=2, output_array=None):
     # does not support the axis keyword. So, we must use a for loop.
     # Take each image at each time and turn the image into a vector.
     # Then, find the norm and divide each image by this norm.
-    for i in xrange(output_array.shape[0]):
+    for i in iters.irange(output_array.shape[0]):
         output_array_i = output_array[i]
         output_array_i_norm = numpy.linalg.norm(
             output_array_i.ravel(), ord=ord