MCPcopy Index your code
hub / github.com/numpy/numpy / test_specific_axes

Method test_specific_axes

numpy/lib/tests/test_function_base.py:1272–1296  ·  view source on GitHub ↗
(self)

Source from the content-addressed store, hash-verified

1270 assert_almost_equal(res1[1], fdx_uneven_ord2)
1271
1272 def test_specific_axes(self):
1273 # Testing that gradient can work on a given axis only
1274 v = [[1, 1], [3, 4]]
1275 x = np.array(v)
1276 dx = [np.array([[2., 3.], [2., 3.]]),
1277 np.array([[0., 0.], [1., 1.]])]
1278 assert_array_equal(gradient(x, axis=0), dx[0])
1279 assert_array_equal(gradient(x, axis=1), dx[1])
1280 assert_array_equal(gradient(x, axis=-1), dx[1])
1281 assert_array_equal(gradient(x, axis=(1, 0)), [dx[1], dx[0]])
1282
1283 # test axis=None which means all axes
1284 assert_almost_equal(gradient(x, axis=None), [dx[0], dx[1]])
1285 # and is the same as no axis keyword given
1286 assert_almost_equal(gradient(x, axis=None), gradient(x))
1287
1288 # test vararg order
1289 assert_array_equal(gradient(x, 2, 3, axis=(1, 0)),
1290 [dx[1] / 2.0, dx[0] / 3.0])
1291 # test maximal number of varargs
1292 assert_raises(TypeError, gradient, x, 1, 2, axis=1)
1293
1294 assert_raises(AxisError, gradient, x, axis=3)
1295 assert_raises(AxisError, gradient, x, axis=-3)
1296 # assert_raises(TypeError, gradient, x, axis=[1,])
1297
1298 def test_timedelta64(self):
1299 # Make sure gradient() can handle special types like timedelta64

Callers

nothing calls this directly

Calls 4

assert_array_equalFunction · 0.90
gradientFunction · 0.90
assert_almost_equalFunction · 0.90
assert_raisesFunction · 0.90

Tested by

no test coverage detected