def test_dygraph(self):
with fluid.dygraph.guard():
np_x = np.random.rand(22, 128, 3).astype('int64')
np_y = np.random.rand(22, 128, 3).astype('int64')
x = paddle.to_tensor(np_x)
y = paddle.to_tensor(np_y)
z = paddle.remainder(x, y)
np_z = z.numpy()
z_expected = np.mod(np_x, np_y)
self.assertEqual((np_z == z_expected).all(), True)
np_x = np.array([-3.3, 11.5, -2, 3.5])
np_y = np.array([-1.2, 2., 3.3, -2.3])
x = paddle.to_tensor(np_x)
y = paddle.to_tensor(np_y)
z = x % y
z_expected = np.array([-0.9, 1.5, 1.3, -1.1])
self.assertEqual(np.allclose(z_expected, z.numpy()), True)
np_x = np.random.rand(22, 128, 3).astype('int32')
np_y = np.random.rand(22, 128, 3).astype('int32')
x = paddle.to_tensor(np_x)
y = paddle.to_tensor(np_y)
z = paddle.remainder(x, y)
np_z = z.numpy()
z_expected = np.mod(np_x, np_y)
self.assertEqual((np_z == z_expected).all(), True)
np_x = np.array([-3, 11, -2, 3])
np_y = np.array([-1, 2, 3, -2])
x = paddle.to_tensor(np_x, dtype="float16")
y = paddle.to_tensor(np_y, dtype="float16")
z = x % y
z_expected = np.array([0, 1, 1, -1])
self.assertEqual(np.allclose(z_expected, z.numpy()), True)
def test_name(self):
with fluid.program_guard(fluid.Program()):
x = fluid.data(name="x", shape=[2, 3], dtype="int64")
y = fluid.data(name='y', shape=[2, 3], dtype='int64')
y_1 = paddle.remainder(x, y, name='div_res')
self.assertEqual(('div_res' in y_1.name), True)
def test_dygraph(self):
paddle.set_device('npu:0')
with fluid.dygraph.guard():
np_x = np.array([2, 3, 8, 7]).astype('int64')
np_y = np.array([1, 5, 3, 3]).astype('int64')
x = paddle.to_tensor(np_x)
y = paddle.to_tensor(np_y)
z = paddle.remainder(x, y)
np_z = z.numpy()
z_expected = np.array([0, 3, 2, 1])
self.assertEqual((np_z == z_expected).all(), True)
np_x = np.array([-3.3, 11.5, -2, 3.5])
np_y = np.array([-1.2, 2., 3.3, -2.3])
x = paddle.to_tensor(np_x)
y = paddle.to_tensor(np_y)
z = x % y
z_expected = np.array([-0.9, 1.5, 1.3, -1.1])
self.assertEqual(np.allclose(z_expected, z.numpy()), True)
np_x = np.array([-3, 11, -2, 3])
np_y = np.array([-1, 2, 3, -2])
x = paddle.to_tensor(np_x, dtype="int64")
y = paddle.to_tensor(np_y, dtype="int64")
z = x % y
z_expected = np.array([0, 1, 1, -1])
self.assertEqual(np.allclose(z_expected, z.numpy()), True)
def _append_optimize_op(self, block, param_and_grad):
one_var = paddle.ones(shape=[1], dtype='int32', name='lookahead_ones')
zero_var = paddle.zeros(shape=[1],
dtype='int32',
name='lookahead_zeros')
k_var = layers.create_global_var(
name=unique_name.generate("lookahead_k"),
shape=[1],
value=self.k,
dtype='int32',
persistable=True)
mod = paddle.remainder(self._global_step_var, k_var)
cond_1 = paddle.equal(self._global_step_var, one_var)
cond_1 = paddle.cast(cond_1, dtype='float32')
cond_2 = paddle.equal(mod, zero_var)
cond_2 = paddle.cast(cond_2, dtype='float32')
slow_var = self._get_accumulator(self._slow_str, param_and_grad[0])
tmp_var = cond_1 * param_and_grad[0] + (1 - cond_1) * slow_var
paddle.assign(tmp_var, slow_var)
tmp_var = self.alpha * param_and_grad[0] + (1.0 -
self.alpha) * slow_var
tmp_var_1 = cond_2 * tmp_var + (1 - cond_2) * param_and_grad[0]
paddle.assign(tmp_var_1, param_and_grad[0])
tmp_var_1 = cond_2 * tmp_var + (1 - cond_2) * slow_var
paddle.assign(tmp_var_1, slow_var)