未验证 提交 2c01c221 编写于 作者: X Xin Pan 提交者: GitHub

Merge pull request #13531 from gongweibao/generator2

Hide kwargs
...@@ -153,6 +153,13 @@ paddle.fluid.layers.elementwise_mul ArgSpec(args=['x', 'y', 'out', 'axis', 'use_ ...@@ -153,6 +153,13 @@ paddle.fluid.layers.elementwise_mul ArgSpec(args=['x', 'y', 'out', 'axis', 'use_
paddle.fluid.layers.elementwise_max ArgSpec(args=['x', 'y', 'out', 'axis', 'use_mkldnn', 'act', 'name'], varargs=None, keywords=None, defaults=(None, -1, False, None, None)) paddle.fluid.layers.elementwise_max ArgSpec(args=['x', 'y', 'out', 'axis', 'use_mkldnn', 'act', 'name'], varargs=None, keywords=None, defaults=(None, -1, False, None, None))
paddle.fluid.layers.elementwise_min ArgSpec(args=['x', 'y', 'out', 'axis', 'use_mkldnn', 'act', 'name'], varargs=None, keywords=None, defaults=(None, -1, False, None, None)) paddle.fluid.layers.elementwise_min ArgSpec(args=['x', 'y', 'out', 'axis', 'use_mkldnn', 'act', 'name'], varargs=None, keywords=None, defaults=(None, -1, False, None, None))
paddle.fluid.layers.elementwise_pow ArgSpec(args=['x', 'y', 'out', 'axis', 'use_mkldnn', 'act', 'name'], varargs=None, keywords=None, defaults=(None, -1, False, None, None)) paddle.fluid.layers.elementwise_pow ArgSpec(args=['x', 'y', 'out', 'axis', 'use_mkldnn', 'act', 'name'], varargs=None, keywords=None, defaults=(None, -1, False, None, None))
paddle.fluid.layers.uniform_random_batch_size_like ArgSpec(args=['input', 'shape', 'dtype', 'input_dim_idx', 'output_dim_idx', 'min', 'max', 'seed'], varargs=None, keywords=None, defaults=('float32', 0, 0, -1.0, 1.0, 0))
paddle.fluid.layers.gaussian_random ArgSpec(args=['shape', 'mean', 'std', 'seed', 'dtype', 'use_mkldnn'], varargs=None, keywords=None, defaults=(0.0, 1.0, 0, 'float32', False))
paddle.fluid.layers.sampling_id ArgSpec(args=['x', 'min', 'max', 'seed', 'dtype'], varargs=None, keywords=None, defaults=(0.0, 1.0, 0, 'float32'))
paddle.fluid.layers.gaussian_random_batch_size_like ArgSpec(args=['input', 'shape', 'input_dim_idx', 'output_dim_idx', 'mean', 'std', 'seed', 'dtype'], varargs=None, keywords=None, defaults=(0, 0, 0.0, 1.0, 0, 'float32'))
paddle.fluid.layers.sum ArgSpec(args=['x', 'use_mkldnn'], varargs=None, keywords=None, defaults=(False,))
paddle.fluid.layers.slice ArgSpec(args=['input', 'axes', 'starts', 'ends'], varargs=None, keywords=None, defaults=None)
paddle.fluid.layers.shape ArgSpec(args=['input'], varargs=None, keywords=None, defaults=None)
paddle.fluid.layers.data ArgSpec(args=['name', 'shape', 'append_batch_size', 'dtype', 'lod_level', 'type', 'stop_gradient'], varargs=None, keywords=None, defaults=(True, 'float32', 0, VarType.LOD_TENSOR, True)) paddle.fluid.layers.data ArgSpec(args=['name', 'shape', 'append_batch_size', 'dtype', 'lod_level', 'type', 'stop_gradient'], varargs=None, keywords=None, defaults=(True, 'float32', 0, VarType.LOD_TENSOR, True))
paddle.fluid.layers.open_files ArgSpec(args=['filenames', 'shapes', 'lod_levels', 'dtypes', 'thread_num', 'buffer_size', 'pass_num', 'is_test'], varargs=None, keywords=None, defaults=(None, None, 1, None)) paddle.fluid.layers.open_files ArgSpec(args=['filenames', 'shapes', 'lod_levels', 'dtypes', 'thread_num', 'buffer_size', 'pass_num', 'is_test'], varargs=None, keywords=None, defaults=(None, None, 1, None))
paddle.fluid.layers.read_file ArgSpec(args=['reader'], varargs=None, keywords=None, defaults=None) paddle.fluid.layers.read_file ArgSpec(args=['reader'], varargs=None, keywords=None, defaults=None)
...@@ -224,13 +231,6 @@ paddle.fluid.layers.logical_and ArgSpec(args=[], varargs='args', keywords='kwarg ...@@ -224,13 +231,6 @@ paddle.fluid.layers.logical_and ArgSpec(args=[], varargs='args', keywords='kwarg
paddle.fluid.layers.logical_or ArgSpec(args=[], varargs='args', keywords='kwargs', defaults=None) paddle.fluid.layers.logical_or ArgSpec(args=[], varargs='args', keywords='kwargs', defaults=None)
paddle.fluid.layers.logical_xor ArgSpec(args=[], varargs='args', keywords='kwargs', defaults=None) paddle.fluid.layers.logical_xor ArgSpec(args=[], varargs='args', keywords='kwargs', defaults=None)
paddle.fluid.layers.logical_not ArgSpec(args=[], varargs='args', keywords='kwargs', defaults=None) paddle.fluid.layers.logical_not ArgSpec(args=[], varargs='args', keywords='kwargs', defaults=None)
paddle.fluid.layers.uniform_random_batch_size_like ArgSpec(args=[], varargs='args', keywords='kwargs', defaults=None)
paddle.fluid.layers.gaussian_random ArgSpec(args=[], varargs='args', keywords='kwargs', defaults=None)
paddle.fluid.layers.sampling_id ArgSpec(args=[], varargs='args', keywords='kwargs', defaults=None)
paddle.fluid.layers.gaussian_random_batch_size_like ArgSpec(args=[], varargs='args', keywords='kwargs', defaults=None)
paddle.fluid.layers.sum ArgSpec(args=[], varargs='args', keywords='kwargs', defaults=None)
paddle.fluid.layers.slice ArgSpec(args=[], varargs='args', keywords='kwargs', defaults=None)
paddle.fluid.layers.shape ArgSpec(args=[], varargs='args', keywords='kwargs', defaults=None)
paddle.fluid.layers.maxout ArgSpec(args=[], varargs='args', keywords='kwargs', defaults=None) paddle.fluid.layers.maxout ArgSpec(args=[], varargs='args', keywords='kwargs', defaults=None)
paddle.fluid.layers.sigmoid ArgSpec(args=['x', 'name'], varargs=None, keywords=None, defaults=(None,)) paddle.fluid.layers.sigmoid ArgSpec(args=['x', 'name'], varargs=None, keywords=None, defaults=(None,))
paddle.fluid.layers.logsigmoid ArgSpec(args=['x', 'name'], varargs=None, keywords=None, defaults=(None,)) paddle.fluid.layers.logsigmoid ArgSpec(args=['x', 'name'], varargs=None, keywords=None, defaults=(None,))
......
...@@ -53,15 +53,16 @@ class SamplingIdOpMaker : public framework::OpProtoAndCheckerMaker { ...@@ -53,15 +53,16 @@ class SamplingIdOpMaker : public framework::OpProtoAndCheckerMaker {
SamplingId Operator. SamplingId Operator.
A layer for sampling id from multinomial distribution from the A layer for sampling id from multinomial distribution from the
input. Sampling one id for one sample.)DOC"); input. Sampling one id for one sample.)DOC");
AddAttr<float>("min", "Minimum value of random. [default 0.0].") AddAttr<float>("min", "Minimum value of random. (float, default 0.0).")
.SetDefault(0.0f); .SetDefault(0.0f);
AddAttr<float>("max", "Maximun value of random. [default 1.0].") AddAttr<float>("max", "Maximun value of random. (float, default 1.0).")
.SetDefault(1.0f); .SetDefault(1.0f);
AddAttr<int>("seed", AddAttr<int>(
"Random seed used for the random number engine. " "seed",
"0 means use a seed generated by the system." "Random seed used for the random number engine. "
"Note that if seed is not 0, this operator will always " "0 means use a seed generated by the system."
"generate the same random numbers every time. [default 0].") "Note that if seed is not 0, this operator will always "
"generate the same random numbers every time. (int, default 0).")
.SetDefault(0); .SetDefault(0);
} }
}; };
......
...@@ -284,7 +284,7 @@ def detection_output(loc, ...@@ -284,7 +284,7 @@ def detection_output(loc,
target_box=loc, target_box=loc,
code_type='decode_center_size') code_type='decode_center_size')
compile_shape = scores.shape compile_shape = scores.shape
run_shape = ops.shape(scores) run_shape = nn.shape(scores)
scores = nn.flatten(x=scores, axis=2) scores = nn.flatten(x=scores, axis=2)
scores = nn.softmax(input=scores) scores = nn.softmax(input=scores)
scores = nn.reshape(x=scores, shape=compile_shape, actual_shape=run_shape) scores = nn.reshape(x=scores, shape=compile_shape, actual_shape=run_shape)
...@@ -697,7 +697,7 @@ def ssd_loss(location, ...@@ -697,7 +697,7 @@ def ssd_loss(location,
raise ValueError("Only support mining_type == max_negative now.") raise ValueError("Only support mining_type == max_negative now.")
num, num_prior, num_class = confidence.shape num, num_prior, num_class = confidence.shape
conf_shape = ops.shape(confidence) conf_shape = nn.shape(confidence)
def __reshape_to_2d(var): def __reshape_to_2d(var):
return nn.flatten(x=var, axis=2) return nn.flatten(x=var, axis=2)
...@@ -724,7 +724,7 @@ def ssd_loss(location, ...@@ -724,7 +724,7 @@ def ssd_loss(location,
target_label.stop_gradient = True target_label.stop_gradient = True
conf_loss = nn.softmax_with_cross_entropy(confidence, target_label) conf_loss = nn.softmax_with_cross_entropy(confidence, target_label)
# 3. Mining hard examples # 3. Mining hard examples
actual_shape = ops.slice(conf_shape, axes=[0], starts=[0], ends=[2]) actual_shape = nn.slice(conf_shape, axes=[0], starts=[0], ends=[2])
actual_shape.stop_gradient = True actual_shape.stop_gradient = True
conf_loss = nn.reshape( conf_loss = nn.reshape(
x=conf_loss, shape=(num, num_prior), actual_shape=actual_shape) x=conf_loss, shape=(num, num_prior), actual_shape=actual_shape)
......
...@@ -29,110 +29,29 @@ from .. import unique_name ...@@ -29,110 +29,29 @@ from .. import unique_name
from functools import reduce from functools import reduce
__all__ = [ __all__ = [
'fc', 'fc', 'embedding', 'dynamic_lstm', 'dynamic_lstmp', 'dynamic_gru',
'embedding', 'gru_unit', 'linear_chain_crf', 'crf_decoding', 'cos_sim', 'cross_entropy',
'dynamic_lstm', 'square_error_cost', 'chunk_eval', 'sequence_conv', 'conv2d', 'conv3d',
'dynamic_lstmp', 'sequence_pool', 'sequence_softmax', 'softmax', 'pool2d', 'pool3d',
'dynamic_gru', 'batch_norm', 'beam_search_decode', 'conv2d_transpose', 'conv3d_transpose',
'gru_unit', 'sequence_expand', 'sequence_expand_as', 'sequence_pad', 'lstm_unit',
'linear_chain_crf', 'reduce_sum', 'reduce_mean', 'reduce_max', 'reduce_min', 'reduce_prod',
'crf_decoding', 'sequence_first_step', 'sequence_last_step', 'dropout', 'split',
'cos_sim', 'ctc_greedy_decoder', 'edit_distance', 'l2_normalize', 'matmul', 'topk',
'cross_entropy', 'warpctc', 'sequence_reshape', 'transpose', 'im2sequence', 'nce',
'square_error_cost', 'hsigmoid', 'beam_search', 'row_conv', 'multiplex', 'layer_norm',
'chunk_eval', 'softmax_with_cross_entropy', 'smooth_l1', 'one_hot',
'sequence_conv', 'autoincreased_step_counter', 'reshape', 'squeeze', 'unsqueeze',
'conv2d', 'lod_reset', 'lrn', 'pad', 'pad_constant_like', 'label_smooth', 'roi_pool',
'conv3d', 'dice_loss', 'image_resize', 'image_resize_short', 'resize_bilinear',
'sequence_pool', 'gather', 'scatter', 'sequence_scatter', 'random_crop', 'mean_iou', 'relu',
'sequence_softmax', 'log', 'crop', 'rank_loss', 'elu', 'relu6', 'pow', 'stanh', 'hard_sigmoid',
'softmax', 'swish', 'prelu', 'brelu', 'leaky_relu', 'soft_relu', 'flatten',
'pool2d', 'sequence_mask', 'stack', 'pad2d', 'unstack', 'sequence_enumerate',
'pool3d', 'expand', 'sequence_concat', 'scale', 'elementwise_add', 'elementwise_div',
'batch_norm', 'elementwise_sub', 'elementwise_mul', 'elementwise_max', 'elementwise_min',
'beam_search_decode', 'elementwise_pow', 'uniform_random_batch_size_like', 'gaussian_random',
'conv2d_transpose', 'sampling_id', 'gaussian_random_batch_size_like', 'sum', 'slice', 'shape'
'conv3d_transpose',
'sequence_expand',
'sequence_expand_as',
'sequence_pad',
'lstm_unit',
'reduce_sum',
'reduce_mean',
'reduce_max',
'reduce_min',
'reduce_prod',
'sequence_first_step',
'sequence_last_step',
'dropout',
'split',
'ctc_greedy_decoder',
'edit_distance',
'l2_normalize',
'matmul',
'topk',
'warpctc',
'sequence_reshape',
'transpose',
'im2sequence',
'nce',
'hsigmoid',
'beam_search',
'row_conv',
'multiplex',
'layer_norm',
'softmax_with_cross_entropy',
'smooth_l1',
'one_hot',
'autoincreased_step_counter',
'reshape',
'squeeze',
'unsqueeze',
'lod_reset',
'lrn',
'pad',
'pad_constant_like',
'label_smooth',
'roi_pool',
'dice_loss',
'image_resize',
'image_resize_short',
'resize_bilinear',
'gather',
'scatter',
'sequence_scatter',
'random_crop',
'mean_iou',
'relu',
'log',
'crop',
'rank_loss',
'elu',
'relu6',
'pow',
'stanh',
'hard_sigmoid',
'swish',
'prelu',
'brelu',
'leaky_relu',
'soft_relu',
'flatten',
'sequence_mask',
'stack',
'pad2d',
'unstack',
'sequence_enumerate',
'expand',
'sequence_concat',
'scale',
'elementwise_add',
'elementwise_div',
'elementwise_sub',
'elementwise_mul',
'elementwise_max',
'elementwise_min',
'elementwise_pow',
] ]
...@@ -6463,6 +6382,246 @@ def expand(x, expand_times, name=None): ...@@ -6463,6 +6382,246 @@ def expand(x, expand_times, name=None):
return out return out
from paddle.fluid.framework import convert_np_dtype_to_dtype_
@templatedoc()
def uniform_random_batch_size_like(input,
shape,
dtype='float32',
input_dim_idx=0,
output_dim_idx=0,
min=-1.0,
max=1.0,
seed=0):
"""
${comment}
Args:
input (Variable): ${input_comment}
shape (tuple|list): ${shape_comment}
input_dim_idx (Int): ${input_dim_idx_comment}
output_dim_idx (Int): ${output_dim_idx_comment}
min (Float): ${min_comment}
max (Float): ${max_comment}
seed (Int): ${seed_comment}
dtype(np.dtype|core.VarDesc.VarType|str): The type of data : float32, float_16, int etc
Returns:
out (Variable): ${out_comment}
"""
helper = LayerHelper('uniform_random_batch_size_like', **locals())
out = helper.create_tmp_variable(dtype)
c_dtype = convert_np_dtype_to_dtype_(dtype)
helper.append_op(
type='uniform_random_batch_size_like',
inputs={'Input': input},
outputs={'Out': out},
attrs={
'shape': shape,
'input_dim_idx': input_dim_idx,
'output_dim_idx': output_dim_idx,
'min': min,
'max': max,
'seed': seed,
'dtype': c_dtype
})
return out
@templatedoc()
def gaussian_random(shape,
mean=0.0,
std=1.0,
seed=0,
dtype='float32',
use_mkldnn=False):
"""
${comment}
Args:
shape (tuple|list): ${shape_comment}
mean (Float): ${mean_comment}
std (Float): ${std_comment}
seed (Int): ${seed_comment}
dtype(np.dtype|core.VarDesc.VarType|str): Output data type.
use_mkldnn (Bool): Only used in mkldnn kernel.
Returns:
out (Variable): ${out_comment}
"""
helper = LayerHelper('gaussian_random', **locals())
out = helper.create_tmp_variable(dtype)
c_dtype = convert_np_dtype_to_dtype_(dtype)
helper.append_op(
type='gaussian_random',
outputs={'Out': out},
attrs={
'shape': shape,
'mean': mean,
'std': std,
'seed': seed,
'dtype': c_dtype,
'use_mkldnn': use_mkldnn
})
return out
@templatedoc()
def sampling_id(x, min=0.0, max=1.0, seed=0, dtype='float32'):
"""
${comment}
Args:
x (Variable): ${x_comment}
min (Float): ${min_comment}
max (Float): ${max_comment}
seed (Float): ${seed_comment}
dtype(np.dtype|core.VarDesc.VarType|str): The type of output data : float32, float_16, int etc
Returns:
out (Variable): ${out_comment}
"""
helper = LayerHelper('sampling_id', **locals())
out = helper.create_tmp_variable(dtype)
helper.append_op(
type='sampling_id',
inputs={'X': x},
outputs={'Out': out},
attrs={'min': min,
'max': max,
'seed': seed})
return out
@templatedoc()
def gaussian_random_batch_size_like(input,
shape,
input_dim_idx=0,
output_dim_idx=0,
mean=0.0,
std=1.0,
seed=0,
dtype='float32'):
"""
${comment}
Args:
input (Variable): ${input_comment}
shape (tuple|list): ${shape_comment}
input_dim_idx (Int): ${input_dim_idx_comment}
output_dim_idx (Int): ${output_dim_idx_comment}
mean (Float): ${mean_comment}
std (Float): ${std_comment}
seed (Int): ${seed_comment}
dtype(np.dtype|core.VarDesc.VarType|str): The type of output data : float32, float_16, int etc
Returns:
out (Variable): ${out_comment}
"""
helper = LayerHelper('gaussian_random_batch_size_like', **locals())
out = helper.create_tmp_variable(dtype)
c_dtype = convert_np_dtype_to_dtype_(dtype)
helper.append_op(
type='gaussian_random_batch_size_like',
inputs={'Input': input},
outputs={'Out': out},
attrs={
'shape': shape,
'input_dim_idx': input_dim_idx,
'output_dim_idx': output_dim_idx,
'mean': mean,
'std': std,
'seed': seed,
'dtype': c_dtype
})
return out
@templatedoc()
def sum(x, use_mkldnn=False):
"""
${comment}
Args:
x (Variable): ${x_comment}
use_mkldnn (Bool): ${use_mkldnn_comment}
Returns:
out (Variable): ${out_comment}
"""
helper = LayerHelper('sum', **locals())
out = helper.create_tmp_variable(dtype=helper.input_dtype('x'))
helper.append_op(
type='sum',
inputs={'X': x},
outputs={'Out': out},
attrs={'use_mkldnn': use_mkldnn})
return out
@templatedoc()
def slice(input, axes, starts, ends):
"""
${comment}
Args:
input (Variable): ${input_comment}.
axes (List): ${axes_comment}
starts (List): ${starts_comment}
ends (List): ${ends_comment}
Returns:
out (Variable): ${out_comment}
"""
helper = LayerHelper('slice', **locals())
out = helper.create_tmp_variable(dtype=helper.input_dtype('input'))
helper.append_op(
type='slice',
inputs={'Input': input},
outputs={'Out': out},
attrs={'axes': axes,
'starts': starts,
'ends': ends})
return out
@templatedoc()
def shape(input):
"""
${comment}
Args:
input (Variable): ${input_comment}
Returns:
out (Variable): ${out_comment}
"""
helper = LayerHelper('shape', **locals())
out = helper.create_tmp_variable(dtype=helper.input_dtype('input'))
helper.append_op(
type='shape', inputs={'Input': input}, outputs={'Out': out})
return out
def _elementwise_op(helper): def _elementwise_op(helper):
op_type = helper.layer_type op_type = helper.layer_type
x = helper.kwargs.get('x', None) x = helper.kwargs.get('x', None)
......
...@@ -45,13 +45,6 @@ __all__ = [ ...@@ -45,13 +45,6 @@ __all__ = [
'logical_or', 'logical_or',
'logical_xor', 'logical_xor',
'logical_not', 'logical_not',
'uniform_random_batch_size_like',
'gaussian_random',
'sampling_id',
'gaussian_random_batch_size_like',
'sum',
'slice',
'shape',
'maxout', 'maxout',
] ]
......
...@@ -541,7 +541,7 @@ class TestBook(unittest.TestCase): ...@@ -541,7 +541,7 @@ class TestBook(unittest.TestCase):
with program_guard(program): with program_guard(program):
input = layers.data( input = layers.data(
name="input", shape=[3, 100, 100], dtype="float32") name="input", shape=[3, 100, 100], dtype="float32")
out = layers.shape(input, name="shape") out = layers.shape(input)
self.assertIsNotNone(out) self.assertIsNotNone(out)
print(str(program)) print(str(program))
...@@ -758,6 +758,65 @@ class TestBook(unittest.TestCase): ...@@ -758,6 +758,65 @@ class TestBook(unittest.TestCase):
out = layers.expand(x, [1, 2]) out = layers.expand(x, [1, 2])
print(str(program)) print(str(program))
def test_uniform_random_batch_size_like(self):
program = Program()
with program_guard(program):
input = layers.data(name="input", shape=[13, 11], dtype='float32')
out = layers.uniform_random_batch_size_like(input, [-1, 11])
self.assertIsNotNone(out)
print(str(program))
def test_gaussian_random(self):
program = Program()
with program_guard(program):
out = layers.gaussian_random(shape=[20, 30])
self.assertIsNotNone(out)
print(str(program))
def test_sampling_id(self):
program = Program()
with program_guard(program):
x = layers.data(
name="X",
shape=[13, 11],
dtype='float32',
append_batch_size=False)
out = layers.sampling_id(x)
self.assertIsNotNone(out)
print(str(program))
def test_gaussian_random_batch_size_like(self):
program = Program()
with program_guard(program):
input = layers.data(name="input", shape=[13, 11], dtype='float32')
out = layers.gaussian_random_batch_size_like(
input, shape=[-1, 11], mean=1.0, std=2.0)
self.assertIsNotNone(out)
print(str(program))
def test_sum(self):
program = Program()
with program_guard(program):
input = layers.data(name="input", shape=[13, 11], dtype='float32')
out = layers.sum(input)
self.assertIsNotNone(out)
print(str(program))
def test_slice(self):
starts = [1, 0, 2]
ends = [3, 3, 4]
axes = [0, 1, 2]
program = Program()
with program_guard(program):
input = layers.data(
name="input", shape=[3, 4, 5, 6], dtype='float32')
out = layers.slice(input, axes=axes, starts=starts, ends=ends)
def test_softshrink(self): def test_softshrink(self):
program = Program() program = Program()
with program_guard(program): with program_guard(program):
......
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册