def test_inflate(self): st = SingleTransform([0, 0, 0, 0, 0, 0]) st.inflate(reshape(matrix([1, 0, 0, 0, 0, 0], float), (-1, 1))) expected = numpy.matrix( [[1, 0, 0, 1], [0, 1, 0, 0], [0, 0, 1, 0], [0, 0, 0, 1]], float) print "" print st.transform self.assertAlmostEqual(numpy.linalg.norm(st.transform - expected), 0.0, 6)
def test_deflate(self): st = SingleTransform([0, 0, 0, 0, 1, 0]) for i in range(5): print "next test_deflate" p = numpy.random.rand(6, 1) st.inflate(p) result = st.deflate() print result print p self.assertAlmostEqual(numpy.linalg.norm(p - result), 0.0, 6)
def test_inflate(self): st = SingleTransform([0, 0, 0, 0, 0, 0]) st.inflate( reshape( matrix([1, 0, 0, 0, 0, 0], float), (-1,1) )) expected = numpy.matrix( [[ 1, 0, 0, 1], [ 0, 1, 0, 0], [ 0, 0, 1, 0], [ 0, 0, 0, 1]], float ) print "" print st.transform self.assertAlmostEqual(numpy.linalg.norm(st.transform-expected), 0.0, 6)
def test_equality(self): st = SingleTransform() for i in range(5): print "------next-------" p = numpy.random.rand(6, 1) new = st.inflate(p, True) old = st.inflate_old(p) print new print old print "diff: " print old - new self.assertAlmostEqual(numpy.linalg.norm(old - new), 0.0, 6)
def test_deflate(self): st = SingleTransform([0, 0, 0, 0, 0, 0]) p = reshape(matrix([1, 0, 0, 0, 0, 0], float), (-1, 1)) st.inflate(p) result = st.deflate() self.assertAlmostEqual(numpy.linalg.norm(p - result), 0.0, 6)
def test_deflate(self): st = SingleTransform([0, 0, 0, 0, 0, 0]) p = reshape( matrix([1, 0, 0, 0, 0, 0], float), (-1,1) ) st.inflate(p) result = st.deflate() self.assertAlmostEqual(numpy.linalg.norm(p-result), 0.0, 6)