Jon Crall Erotemic

Initial work for updating fletch to target python3.

I still haven't gotten this to build all the way through yet, so for now its just a work in progress.

This combines PRs #172 #173 #174 #175 #177 and #186 which each pertain to their own library --- opencv, caffe (which is in there twice), vtk, ffmpeg, and openblas --- and then adds initial support for python3, which is used by multiple libraries.

Current Status:

This branch merges the branches from the other PRs I've recently submitted and then adds initial support for python3 on top. By default only one package (OpenBLAS) has changed versions because it seems that this update never caused any issues in my tests. The other packages still hicup every so often when switching to their new version (some more than others).

	# -- coding: utf-8 --
	from __future__ import absolute_import, division, print_function, unicode_literals
	import numpy as np
	from six.moves import zip

	def toydata1(rng):
	from sklearn.datasets import samples_generator
	n_samples = 500
	n_features = 2
	n_classes = 2

	# -- coding: utf-8 --
	from __future__ import print_function, division, absolute_import, unicode_literals
	import time
	import itertools as it
	from sklearn.utils.extmath import row_norms
	from sklearn.metrics.pairwise import euclidean_distances
	from sklearn.metrics.pairwise import pairwise_distances_argmin_min
	import numpy as np

	# -- coding: utf-8 --
	from __future__ import print_function, division, absolute_import, unicode_literals
	import time
	import itertools as it
	from sklearn.utils.extmath import row_norms
	from sklearn.metrics.pairwise import euclidean_distances
	from sklearn.metrics.pairwise import pairwise_distances_argmin_min
	import numpy as np

	# -- coding: utf-8 --
	from __future__ import print_function, division, absolute_import, unicode_literals
	import time
	import itertools as it
	from sklearn.utils.extmath import row_norms
	from sklearn.metrics.pairwise import euclidean_distances
	from sklearn.metrics.pairwise import pairwise_distances_argmin_min
	import numpy as np

	from scipy.stats import hypergeom

	deck_size = 30
	hand_size = 4

	def mul_for_card(copies_in_deck=2, hand_size=3):
	deck_size = 30
	prb = hypergeom(deck_size, copies_in_deck, hand_size)
	# P(initial_miss)
	p_none_in_first3 = prb.cdf(0)

	import numpy as np
	import utool as ut


	def get_toydata(rng):
	if ut.get_argflag('--toy2'):
	X_true, X, y = toydata2(rng)
	else:
	X_true, X, y = toydata1(rng)
	return X_true, X, y

	# -- coding: utf-8 --
	"""
	Algorithms for finding k-edge-connected compoments.
	"""
	import networkx as nx
	from networkx.utils import arbitrary_element


	class EdgeComponentAuxGraph(object):
	"""

	import numpy as np
	from sklearn.metrics.classification import (unique_labels, _check_targets,
	check_consistent_length,
	coo_matrix)


	def confusion_matrix_orig(y_true, y_pred, labels=None, sample_weight=None):
	y_type, y_true, y_pred = _check_targets(y_true, y_pred)
	if y_type not in ("binary", "multiclass"):
	raise ValueError("%s is not supported" % y_type)