def test_constant_initializer_default_value(self):
"""Test the constant initializer with default value
"""
program = framework.Program()
block = program.global_block()
block.create_parameter(dtype="float32",
shape=[5, 10],
lod_level=0,
name="param",
initializer=initializer.ConstantInitializer())
self.assertEqual(len(block.ops), 1)
init_op = block.ops[0]
self.assertEqual(init_op.type, 'fill_constant')
self.assertAlmostEqual(init_op.attr('value'), 0.0, delta=DELTA)
def paddle_random_normal(shape, loc=.0, scale=1., seed=1, dtype="float32"):
program = framework.Program()
block = program.global_block()
w = block.create_var(
dtype="float32",
shape=shape,
lod_level=0,
name="param",
initializer=initializer.NormalInitializer(
loc=.0, scale=scale, seed=seed))
place = core.CPUPlace()
exe = Executor(place)
out = exe.run(program, fetch_list=[w])
return np.array(out[0])
def test_uniform_initializer_default_value(self):
"""Test the uniform initializer with default value
"""
program = framework.Program()
block = program.global_block()
block.create_parameter(dtype="float32",
shape=[5, 10],
lod_level=0,
name="param",
initializer=initializer.UniformInitializer())
self.assertEqual(len(block.ops), 1)
init_op = block.ops[0]
self.assertEqual(init_op.type, 'uniform_random')
self.assertAlmostEqual(init_op.attr('min'), -1.0, delta=DELTA)
self.assertAlmostEqual(init_op.attr('max'), 1.0, delta=DELTA)
self.assertEqual(init_op.attr('seed'), 0)
def test_uniform_initializer(self):
"""Test uniform initializer with supplied attributes
"""
program = framework.Program()
block = program.global_block()
block.create_parameter(dtype="float32",
shape=[5, 10],
lod_level=0,
name="param",
initializer=initializer.UniformInitializer(
-4.2, 3.1, 123))
self.assertEqual(len(block.ops), 1)
init_op = block.ops[0]
self.assertEqual(init_op.type, 'uniform_random')
self.assertAlmostEqual(init_op.attr('min'), -4.2, delta=DELTA)
self.assertAlmostEqual(init_op.attr('max'), 3.1, delta=DELTA)
self.assertEqual(init_op.attr('seed'), 123)
def test_normal_initializer(self):
"""Test normal initializer with supplied attributes
"""
program = framework.Program()
block = program.global_block()
block.create_parameter(dtype="float32",
shape=[5, 10],
lod_level=0,
name="param",
initializer=initializer.NormalInitializer(
2.3, 1.9, 123))
self.assertEqual(len(block.ops), 1)
init_op = block.ops[0]
self.assertEqual(init_op.type, 'gaussian_random')
self.assertAlmostEqual(init_op.attr('mean'), 2.3, delta=DELTA)
self.assertAlmostEqual(init_op.attr('std'), 1.9, delta=DELTA)
self.assertEqual(init_op.attr('seed'), 123)
def test_msra_initializer_supplied_arguments(self):
"""Test the MSRA initializer with supplied arguments
"""
program = framework.Program()
block = program.global_block()
block.create_parameter(dtype="float32",
shape=[5, 10],
lod_level=0,
name="param",
initializer=initializer.MSRAInitializer(
fan_in=12, seed=134))
self.assertEqual(len(block.ops), 1)
init_op = block.ops[0]
self.assertEqual(init_op.type, 'uniform_random')
limit = np.sqrt(6.0 / 12)
self.assertAlmostEqual(init_op.attr('min'), -limit, delta=DELTA)
self.assertAlmostEqual(init_op.attr('max'), limit, delta=DELTA)
self.assertEqual(init_op.attr('seed'), 134)
def test_l2decay_regularizer(self):
program = framework.Program()
block = program.global_block()
mul_x = block.create_parameter(
dtype="float32",
shape=[5, 10],
lod_level=0,
name="mul.x",
regularizer=regularizer.L1DecayRegularizer(0.5))
self.assertTrue(mul_x.regularizer is not None)
self.assertTrue(
isinstance(mul_x.regularizer, regularizer.L1DecayRegularizer))
mul_y = block.create_var(dtype="float32",
shape=[10, 8],
lod_level=0,
name="mul.y")
mul_out = block.create_var(dtype="float32",
shape=[5, 8],
lod_level=0,
name="mul.out")
block.append_op(type="mul",
inputs={
"X": mul_x,
"Y": mul_y
},
outputs={"Out": mul_out},
attrs={"x_num_col_dims": 1})
mean_out = block.create_var(dtype="float32",
shape=[1],
lod_level=0,
name="mean.out")
block.append_op(type="mean",
inputs={"X": mul_out},
outputs={"Out": mean_out})
params_grads = append_backward_ops(mean_out)
self.assertEqual(len(params_grads), 1)
count_ops = len(block.ops)
params_grads = optimizer.append_regularization_ops(params_grads)
self.assertEqual(len(params_grads), 1)
self.assertEqual(len(block.ops), count_ops + 3)
self.assertEqual(block.ops[-1].type, 'elementwise_add')
self.assertEqual(block.ops[-2].type, 'scale')
self.assertEqual(block.ops[-3].type, 'sign')
def test_normal_msra_initializer(self):
"""Test MSRA initializer with normal distribution on
for matrix multiply.
"""
program = framework.Program()
block = program.global_block()
param = block.create_parameter(
dtype="float32",
shape=[5, 10],
lod_level=0,
name="param",
initializer=initializer.MSRAInitializer(uniform=False))
self.assertEqual(len(block.ops), 1)
init_op = block.ops[0]
self.assertEqual(init_op.type, 'gaussian_random')
std = np.sqrt(2.0 / param.shape[0])
self.assertAlmostEqual(init_op.attr('mean'), 0.0, delta=DELTA)
self.assertAlmostEqual(init_op.attr('std'), std, delta=DELTA)
self.assertEqual(init_op.attr('seed'), 0)
def test_uniform_msra_initializer(self):
"""Test MSRA initializer with uniform distribution on
for matrix multiply.
"""
program = framework.Program()
block = program.global_block()
param = block.create_parameter(
dtype="float32",
shape=[5, 10],
lod_level=0,
name="param",
initializer=initializer.MSRAInitializer())
self.assertEqual(len(block.ops), 1)
init_op = block.ops[0]
self.assertEqual(init_op.type, 'uniform_random')
limit = np.sqrt(6.0 / param.shape[0])
self.assertAlmostEqual(init_op.attr('min'), -limit, delta=DELTA)
self.assertAlmostEqual(init_op.attr('max'), limit, delta=DELTA)
self.assertEqual(init_op.attr('seed'), 0)
def test_uniform_msra_initializer_conv(self):
"""Test MSRA initializer with uniform distribution on
for convolutions.
"""
program = framework.Program()
block = program.global_block()
param = block.create_parameter(
dtype="float32",
shape=[5, 10, 15, 20],
lod_level=0,
name="param",
initializer=initializer.MSRAInitializer())
self.assertEqual(len(block.ops), 1)
init_op = block.ops[0]
self.assertEqual(init_op.type, 'uniform_random')
receptive_field_size = float(15 * 20)
limit = np.sqrt(6.0 / (param.shape[1] * receptive_field_size))
self.assertAlmostEqual(init_op.attr('min'), -limit, delta=DELTA)
self.assertAlmostEqual(init_op.attr('max'), limit, delta=DELTA)
self.assertEqual(init_op.attr('seed'), 0)
def test_uniform_initializer_random_seed(self):
"""Test the uniform initializer with manually setting seed
"""
program = framework.Program()
program.random_seed = 123
block = program.global_block()
block.create_parameter(dtype="float32",
shape=[5, 10],
lod_level=0,
name="param",
initializer=initializer.UniformInitializer())
block.create_parameter(
dtype="float32",
shape=[5, 10],
lod_level=0,
name="param",
initializer=initializer.UniformInitializer(seed=456))
init_op = block.ops[1]
self.assertEqual(init_op.attr("seed"), 123)
init_op1 = block.ops[0]
self.assertEqual(init_op1.attr("seed"), 456)
def test_normal_xavier_initializer_conv(self):
"""Test Xavier initializer with normal distribution on
for convolutions.
"""
program = framework.Program()
block = program.global_block()
param = block.create_parameter(
dtype="float32",
shape=[5, 10, 15, 20],
lod_level=0,
name="param",
initializer=initializer.XavierInitializer(uniform=False))
self.assertEqual(len(block.ops), 1)
init_op = block.ops[0]
self.assertEqual(init_op.type, 'gaussian_random')
receptive_field_size = float(15 * 20)
std = np.sqrt(
2.0 / ((param.shape[0] + param.shape[1]) * receptive_field_size))
self.assertAlmostEqual(init_op.attr('mean'), 0.0, delta=DELTA)
self.assertAlmostEqual(init_op.attr('std'), std, delta=DELTA)
self.assertEqual(init_op.attr('seed'), 0)
