def test_fconv_is_faster_than_convolve(): x = np.random.random((6000)) y = np.random.random((6000)) t1 = time.time() np.convolve(x,y) t2 = time.time() fconv(x,y) t3 = time.time() assert_true((t3-t2)<(t2-t1))
def test_fconv_is_faster_than_convolve(): x = np.random.random((6000)) y = np.random.random((6000)) t1 = time.time() np.convolve(x, y) t2 = time.time() fconv(x, y) t3 = time.time() assert_true((t3 - t2) < (t2 - t1))
def test_fconv_returns_double_when_input_is_double(): x = np.random.random((10)).astype(np.double) y = np.random.random((10)).astype(np.double) res = fconv(x,y) assert_true(res.dtype==np.double)
def test_fconv_returns_complex_when_input_is_complex(): x = np.random.random((10)).astype(np.complex) y = np.random.random((10)).astype(np.complex) res = fconv(x,y) assert_true(res.dtype==np.complex)
def _assert_numpy_convolve_gives_same_results_as_fconv_for(x,y): conv_np = np.convolve(x,y) conv_fconv = fconv(x,y) assert_array_almost_equal(conv_np, conv_fconv)
def test_fconv_returns_double_when_input_is_double(): x = np.random.random((10)).astype(np.double) y = np.random.random((10)).astype(np.double) res = fconv(x, y) assert_true(res.dtype == np.double)
def test_fconv_returns_complex_when_input_is_complex(): x = np.random.random((10)).astype(np.complex) y = np.random.random((10)).astype(np.complex) res = fconv(x, y) assert_true(res.dtype == np.complex)
def _assert_numpy_convolve_gives_same_results_as_fconv_for(x, y): conv_np = np.convolve(x, y) conv_fconv = fconv(x, y) assert_array_almost_equal(conv_np, conv_fconv)