Skip to content

P
Paddle

机器未来 / Paddle
与 Fork 源项目一致

Fork自 PaddlePaddle / Paddle

通知 1
Star 1
Fork 0
P
Paddle
提交 b77c886e 编写于 作者: Q qiaolongfei
浏览文件
操作

Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into update-api-reference-1

上级 82a4cf19 e6654c1c
隐藏空白更改
内联 并排
Showing with 517 addition and 175 deletion
paddle/fluid/operators/activation_op.cc
浏览文件 @ b77c886e
......@@ -443,7 +443,7 @@ class SwishOpMaker : public framework::OpProtoAndCheckerMaker {
AddComment(R"DOC(
Swish Activation Operator.
$$out = \frac{x}{1 + e^{- \beta x}}$$
$$out = \\frac{x}{1 + e^{- \beta x}}$$
)DOC");
}
......
paddle/fluid/operators/clip_by_norm_op.cc
浏览文件 @ b77c886e
......@@ -54,10 +54,19 @@ be linearly scaled to make the L2 norm of $Out$ equal to $max\_norm$, as
shown in the following formula:
$$
Out = \frac{max\_norm * X}{norm(X)},
Out = \\frac{max\\_norm * X}{norm(X)},
$$
where $norm(X)$ represents the L2 norm of $X$.
Examples:
.. code-block:: python
data = fluid.layer.data(
name='data', shape=[2, 4, 6], dtype='float32')
reshaped = fluid.layers.clip_by_norm(
x=data, max_norm=0.5)
)DOC");
}
};
......
paddle/fluid/operators/pool_op.cc
浏览文件 @ b77c886e
......@@ -204,8 +204,6 @@ void Pool2dOpMaker::Make() {
// TODO(dzhwinter): need to registered layout transform function
AddComment(R"DOC(
Pool2d Operator.
The pooling2d operation calculates the output based on
the input, pooling_type and ksize, strides, paddings parameters.
Input(X) and output(Out) are in NCHW format, where N is batch size, C is the
......@@ -215,19 +213,28 @@ These two elements represent height and width, respectively.
The input(X) size and output(Out) size may be different.
Example:
Input:
X shape: $(N, C, H_{in}, W_{in})$
Output:
Out shape: $(N, C, H_{out}, W_{out})$
For ceil_mode = false:
$$
H_{out} = \frac{(H_{in} - ksize[0] + 2 * paddings[0])}{strides[0]} + 1 \\
W_{out} = \frac{(W_{in} - ksize[1] + 2 * paddings[1])}{strides[1]} + 1
H_{out} = \\frac{(H_{in} - ksize[0] + 2 * paddings[0])}{strides[0]} + 1
$$
$$
W_{out} = \\frac{(W_{in} - ksize[1] + 2 * paddings[1])}{strides[1]} + 1
$$
For ceil_mode = true:
$$
H_{out} = \frac{(H_{in} - ksize[0] + 2 * paddings[0] + strides[0] - 1)}{strides[0]} + 1 \\
W_{out} = \frac{(W_{in} - ksize[1] + 2 * paddings[1] + strides[1] - 1)}{strides[1]} + 1
H_{out} = \\frac{(H_{in} - ksize[0] + 2 * paddings[0] + strides[0] - 1)}{strides[0]} + 1
$$
$$
W_{out} = \\frac{(W_{in} - ksize[1] + 2 * paddings[1] + strides[1] - 1)}{strides[1]} + 1
$$
)DOC");
......
paddle/fluid/operators/uniform_random_batch_size_like_op.cc
浏览文件 @ b77c886e
......@@ -35,10 +35,10 @@ class UniformRandomBatchSizeLikeOpMaker : public BatchSizeLikeOpMaker {
protected:
void Apply() override {
AddComment(R"DOC(
Uniform random operator
UniformRandomBatchSizeLike operator.
This operator initializes a tensor with the same batch_size as the Input tensor
with random values sampled from a uniform distribution.
with random values sampled from a uniform distribution.
)DOC");
AddAttr<float>("min",
......
python/paddle/fluid/framework.py
浏览文件 @ b77c886e
......@@ -1034,6 +1034,37 @@ class Block(object):
class Program(object):
"""
Python Program. Beneath it is a ProgramDesc, which is used for
create c++ Program. A program is a self-contained programing
language like container. It has at least one Block, when the
control flow op like conditional_block, while_op is included,
it will contains nested block.
Please reference the framework.proto for details.
Notes: we have default_startup_program and default_main_program
by default, a pair of them will shared the parameters.
The default_startup_program only run once to initialize parameters,
default_main_program run in every minibatch and adjust the weights.
Args:
None
Returns:
Python Program
Examples:
.. code-block:: python
main_program = Program()
startup_program = Program()
with fluid.program_guard(main_program=main_program, startup_program=startup_program):
fluid.layers.data(name="x", shape=[-1, 784], dtype='float32')
fluid.layers.data(name="y", shape=[-1, 1], dtype='int32')
fluid.layers.fc(name="fc", shape=[10], dtype='float32', act="relu")
"""
def __init__(self):
self.desc = core.ProgramDesc()
self.blocks = [Block(self, 0)]
......@@ -1099,6 +1130,8 @@ class Program(object):
def clone(self, for_test=False):
"""Clone the Program object
Args:
for_test(bool): indicate whether clone for test.
Set for_test to False when we want to clone the program for training.
Set for_test to True when we want to clone the program for testing.
......@@ -1109,8 +1142,9 @@ class Program(object):
the is_test attributes in these operators will be set to True for
testing purposes, otherwise, they remain unchanged.
Returns(Program):
The cloned Program object.
Returns:
Program: The cloned Program object.
"""
if for_test:
p = self.inference_optimize()
......@@ -1228,6 +1262,7 @@ class Program(object):
def copy_param_info_from(self, other):
"""
Copy the information of parameters from other program.
Args:
other(Program): Other program
......@@ -1246,6 +1281,7 @@ class Program(object):
def copy_data_info_from(self, other):
"""
Copy the information of data variables from other program.
Args:
other(Program): Other program
......@@ -1299,6 +1335,7 @@ class Parameter(Variable):
def to_string(self, throw_on_error, with_details=False):
"""
To debug string.
Args:
throw_on_error(bool): raise exception when self is not initialized
when throw_on_error is True
......
python/paddle/fluid/layers/control_flow.py
浏览文件 @ b77c886e
......@@ -822,17 +822,25 @@ def max_sequence_len(rank_table):
def lod_tensor_to_array(x, table):
""" Convert a LOD_TENSOR to an LOD_TENSOR_ARRAY.
"""
Convert a LoDTensor to a LoDTensorArray.
This function split a LoDTesnor to a LoDTensorArray according to its LoD
information. LoDTensorArray is an alias of C++ std::vector<LoDTensor> in
PaddlePaddle. The generated LoDTensorArray of this function can be further read
or written by `read_from_array()` and `write_to_array()` operators. However,
this function is generally an internal component of PaddlePaddle `DynamicRNN`.
Users should not use it directly.
Args:
x (Variable|list): The LOD tensor to be converted to a LOD tensor array.
x (Variable|list): The LoDTensor to be converted to a LoDTensorArray.
table (ParamAttr|list): The variable that stores the level of lod
which is ordered by sequence length in
descending order.
descending order. It is generally generated
by `layers.lod_rank_table()` API.
Returns:
Variable: The variable of type array that has been converted from a
tensor.
Variable: The LoDTensorArray that has been converted from the input tensor.
Examples:
.. code-block:: python
......@@ -897,8 +905,7 @@ def increment(x, value=1.0, in_place=True):
in_place (bool): If the increment should be performed in-place.
Returns:
Variable: The tensor variable storing the transformation of
element-wise increment of each value in the input.
Variable: The elementwise-incremented object.
Examples:
.. code-block:: python
......@@ -940,7 +947,7 @@ def array_write(x, i, array=None):
Variable: The output LOD_TENSOR_ARRAY where the input tensor is written.
Examples:
.. code-block::python
.. code-block:: python
tmp = fluid.layers.zeros(shape=[10], dtype='int32')
i = fluid.layers.fill_constant(shape=[1], dtype='int64', value=10)
......@@ -1054,14 +1061,31 @@ def equal(x, y, cond=None, **ignored):
def array_read(array, i):
"""This function performs the operation to read the data in as an
"""
This function performs the operation to read the data in as an
LOD_TENSOR_ARRAY.
.. code-block:: text
Given:
array = [0.6, 0.1, 0.3, 0.1]
And:
i = 2
Then:
output = 0.3
Args:
array (Variable|list): The input tensor that will be written to an array.
i (Variable|list): The subscript index in tensor array, that points the
place where data will be written to.
array (Variable|list): The input tensor that store data to be read.
i (Variable|list): The index of the data to be read from input array.
Returns:
Variable: The tensor type variable that has the data written to it.
Examples:
.. code-block:: python
......@@ -1154,6 +1178,13 @@ def array_length(array):
class ConditionalBlockGuard(BlockGuard):
"""
ConditionalBlockGuard is derived from BlockGuard. It is dedicated for
holding a ConditionalBlock, and helping users entering and exiting the
ConditionalBlock via Python's 'with' keyword. However, ConditionalBlockGuard
is generally an internal component of IfElse, users should not use it directly.
"""
def __init__(self, block):
if not isinstance(block, ConditionalBlock):
raise TypeError("block should be conditional block")
......@@ -1875,26 +1906,26 @@ def reorder_lod_tensor_by_rank(x, rank_table):
def is_empty(x, cond=None, **ignored):
"""
**Is Empty**
This layer returns the truth value of whether the variable is empty.
Test whether a Variable is empty.
Args:
x(Variable): Operand of *is_empty*
cond(Variable|None): Optional output variable to store the result
of *is_empty*
x (Variable): The Variable to be tested.
cond (Variable|None): Output parameter. Returns the test result
of given 'x'. Default: None
Returns:
Variable: The tensor variable storing the output of *is_empty*.
Variable: A bool scalar. True if 'x' is an empty Variable.
Raises:
TypeError: If input cond is not a variable, or cond's dtype is
not bool
not bool.
Examples:
.. code-block:: python
less = fluid.layers.is_empty(x=input)
res = fluid.layers.is_empty(x=input)
# or:
fluid.layers.is_empty(x=input, cond=res)
"""
helper = LayerHelper("is_empty", **locals())
if cond is None:
......
python/paddle/fluid/layers/io.py
浏览文件 @ b77c886e
......@@ -544,6 +544,41 @@ def shuffle(reader, buffer_size):
def batch(reader, batch_size):
"""
This layer is a reader decorator. It takes a reader and adds
'batching' decoration on it. When reading with the result
decorated reader, output data will be automatically organized
to the form of batches.
Args:
reader(Variable): The reader to be decorated with 'batching'.
batch_size(int): The batch size.
Returns:
Variable: The reader which has been decorated with 'batching'.
Examples:
.. code-block:: python
raw_reader = fluid.layers.io.open_files(filenames=['./data1.recordio',
'./data2.recordio'],
shapes=[(3,224,224), (1)],
lod_levels=[0, 0],
dtypes=['float32', 'int64'],
thread_num=2,
buffer_size=2)
batch_reader = fluid.layers.batch(reader=raw_reader, batch_size=5)
# If we read data with the raw_reader:
# data = fluid.layers.read_file(raw_reader)
# We can only get data instance by instance.
#
# However, if we read data with the batch_reader:
# data = fluid.layers.read_file(batch_reader)
# Each 5 adjacent instances will be automatically combined together
# to become a batch. So what we get('data') is a batch data instead
# of an instance.
"""
return __create_unshared_decorated_reader__(
'create_batch_reader', reader, {'batch_size': int(batch_size)})
......@@ -589,15 +624,41 @@ def parallel(reader):
{})
def read_file(file_obj):
def read_file(reader):
"""
Execute the given reader and get data via it.
A reader is also a Variable. It can be a raw reader generated by
`fluid.layers.open_files()` or a decorated one generated by
`fluid.layers.double_buffer()` and so on.
Args:
reader(Variable): The reader to execute.
Returns:
Tuple[Variable]: Data read via the given reader.
Examples:
.. code-block:: python
data_file = fluid.layers.open_files(
filenames=['mnist.recordio'],
shapes=[(-1, 748), (-1, 1)],
lod_levels=[0, 0],
dtypes=["float32", "int64"])
data_file = fluid.layers.double_buffer(
fluid.layers.batch(data_file, batch_size=64))
input, label = fluid.layers.read_file(data_file)
"""
helper = LayerHelper('read_file')
out = [
helper.create_tmp_variable(
stop_gradient=True, dtype='float32')
for _ in range(len(file_obj.desc.shapes()))
for _ in range(len(reader.desc.shapes()))
]
helper.append_op(
type='read', inputs={'Reader': [file_obj]}, outputs={'Out': out})
type='read', inputs={'Reader': [reader]}, outputs={'Out': out})
if len(out) == 1:
return out[0]
else:
......
python/paddle/fluid/layers/learning_rate_scheduler.py
浏览文件 @ b77c886e
......@@ -71,21 +71,40 @@ def noam_decay(d_model, warmup_steps):
def exponential_decay(learning_rate, decay_steps, decay_rate, staircase=False):
"""Applies exponential decay to the learning rate.
"""
Applies exponential decay to the learning rate.
When training a model, it is often recommended to lower the learning rate as the
training progresses. By using this function, the learning rate will be decayed by
'decay_rate' every 'decay_steps' steps.
>>> if staircase == True:
>>> decayed_learning_rate = learning_rate * decay_rate ^ floor(global_step / decay_steps)
>>> else:
>>> decayed_learning_rate = learning_rate * decay_rate ^ (global_step / decay_steps)
```python
decayed_learning_rate = learning_rate *
decay_rate ^ (global_step / decay_steps)
```
Args:
learning_rate: A scalar float32 value or a Variable. This
will be the initial learning rate during training
decay_steps: A Python `int32` number.
decay_rate: A Python `float` number.
staircase: Boolean. If set true, decay the learning rate every decay_steps.
learning_rate(Variable|float): The initial learning rate.
decay_steps(int): See the decay computation above.
decay_rate(float): The decay rate. See the decay computation above.
staircase(Boolean): If True, decay the learning rate at discrete intervals.
Default: False
Returns:
The decayed learning rate
Variable: The decayed learning rate
Examples:
.. code-block:: python
base_lr = 0.1
sgd_optimizer = fluid.optimizer.SGD(
learning_rate=fluid.layers.exponential_decay(
learning_rate=base_lr,
decay_steps=10000,
decay_rate=0.5,
staircase=True))
sgd_optimizer.minimize(avg_cost)
"""
global_step = _decay_step_counter()
......@@ -129,22 +148,39 @@ def natural_exp_decay(learning_rate, decay_steps, decay_rate, staircase=False):
def inverse_time_decay(learning_rate, decay_steps, decay_rate, staircase=False):
"""Applies inverse time decay to the initial learning rate.
"""
Applies inverse time decay to the initial learning rate.
>>> if staircase:
When training a model, it is often recommended to lower the learning rate as the
training progresses. By using this function, an inverse decay function will be
applied to the initial learning rate.
>>> if staircase == True:
>>> decayed_learning_rate = learning_rate / (1 + decay_rate * floor(global_step / decay_step))
>>> else:
>>> decayed_learning_rate = learning_rate / (1 + decay_rate * global_step / decay_step)
Args:
learning_rate: A scalar float32 value or a Variable. This
will be the initial learning rate during training.
decay_steps: A Python `int32` number.
decay_rate: A Python `float` number.
staircase: Boolean. If set true, decay the learning rate every decay_steps.
learning_rate(Variable|float): The initial learning rate.
decay_steps(int): See the decay computation above.
decay_rate(float): The decay rate. See the decay computation above.
staircase(Boolean): If True, decay the learning rate at discrete intervals.
Default: False
Returns:
The decayed learning rate
Variable: The decayed learning rate
Examples:
.. code-block:: python
base_lr = 0.1
sgd_optimizer = fluid.optimizer.SGD(
learning_rate=fluid.layers.inverse_time_decay(
learning_rate=base_lr,
decay_steps=10000,
decay_rate=0.5,
staircase=True))
sgd_optimizer.minimize(avg_cost)
"""
global_step = _decay_step_counter()
......
python/paddle/fluid/layers/metric.py
浏览文件 @ b77c886e
......@@ -27,8 +27,32 @@ __all__ = ['accuracy', 'auc']
def accuracy(input, label, k=1, correct=None, total=None):
"""
accuracy layer.
Refer to the https://en.wikipedia.org/wiki/Precision_and_recall
This function computes the accuracy using the input and label.
The output is the top k inputs and their indices.
If the correct label occurs in top k predictions, then correct will increment by one.
Note: the dtype of accuracy is determined by input. the input and label dtype can be different.
Args:
input(Variable): The input of accuracy layer, which is the predictions of network.
Carry LoD information is supported.
label(Variable): The label of dataset.
k(int): The top k predictions for each class will be checked.
correct(Variable): The correct predictions count.
total(Variable): The total entries count.
Returns:
Variable: The correct rate.
Examples:
.. code-block:: python
data = fluid.layers.data(name="data", shape=[-1, 32, 32], dtype="float32")
label = fluid.layers.data(name="data", shape=[-1,1], dtype="int32")
predict = fluid.layers.fc(input=data, size=10)
acc = fluid.layers.accuracy(input=predict, label=label, k=5)
"""
helper = LayerHelper("accuracy", **locals())
topk_out, topk_indices = nn.topk(input, k=k)
......
python/paddle/fluid/layers/nn.py
浏览文件 @ b77c886e
......@@ -91,6 +91,8 @@ __all__ = [
'gather',
'random_crop',
'mean_iou',
'relu',
'log',
]
......@@ -106,14 +108,15 @@ def fc(input,
"""
**Fully Connected Layer**
The fully connected layer can take multiple tensors as its inputs. It
creates a variable called weights for each input tensor, which represents
a fully connected weight matrix from each input unit to each output unit.
The fully connected layer multiplies each input tensor with its coresponding
weight to produce an output Tensor. If multiple input tensors are given,
the results of multiple multiplications will be sumed up. If bias_attr is
not None, a bias variable will be created and added to the output. Finally,
if activation is not None, it will be applied to the output as well.
This function creates a fully connected layer in the network. It can take
multiple tensors as its inputs. It creates a variable called weights for
each input tensor, which represents a fully connected weight matrix from
each input unit to each output unit. The fully connected layer multiplies
each input tensor with its coresponding weight to produce an output Tensor.
If multiple input tensors are given, the results of multiple multiplications
will be sumed up. If bias_attr is not None, a bias variable will be created
and added to the output. Finally, if activation is not None, it will be applied
to the output as well.
This process can be formulated as follows:
......@@ -154,7 +157,7 @@ def fc(input,
name (str, default None): The name of this layer.
Returns:
A tensor variable storing the transformation result.
Variable: The transformation result.
Raises:
ValueError: If rank of the input tensor is less than 2.
......@@ -162,8 +165,7 @@ def fc(input,
Examples:
.. code-block:: python
data = fluid.layers.data(
name="data", shape=[32, 32], dtype="float32")
data = fluid.layers.data(name="data", shape=[32, 32], dtype="float32")
fc = fluid.layers.fc(input=data, size=1000, act="tanh")
"""
......@@ -845,11 +847,14 @@ def linear_chain_crf(input, label, param_attr=None):
Args:
input(${emission_type}): ${emission_comment}
input(${transition_type}): ${transition_comment}
label(${label_type}): ${label_comment}
param_attr(ParamAttr): The attribute of the learnable parameter.
Returns:
${log_likelihood_comment}
output(${emission_exps_type}): ${emission_exps_comment} \n
output(${transition_exps_type}): ${transition_exps_comment} \n
output(${log_likelihood_type}): ${log_likelihood_comment}
"""
helper = LayerHelper('linear_chain_crf', **locals())
......@@ -911,7 +916,7 @@ def cos_sim(X, Y):
Args:
X (Variable): The input X.
Y (Variable): The input Y.
Returns:
Variable: the output of cosine(X, Y).
"""
......@@ -1117,7 +1122,7 @@ def chunk_eval(input,
chunk_scheme (str): ${chunk_scheme_comment}
num_chunk_types (int): ${num_chunk_types_comment}
excluded_chunk_types (list): ${excluded_chunk_types_comment}
Returns:
tuple: tuple containing: (precision, recall, f1_score,
num_infer_chunks, num_label_chunks,
......@@ -1177,15 +1182,11 @@ def sequence_conv(input,
bias_attr (ParamAttr|None): attributes for bias
param_attr (ParamAttr|None): attributes for parameter
act (str): the activation type
Returns:
Variable: output of sequence_conv
"""
# FIXME(dzh) : want to unify the argument of python layer
# function. So we ignore some unecessary attributes.
# such as, padding_trainable, context_start.
helper = LayerHelper('sequence_conv', **locals())
dtype = helper.input_dtype()
filter_shape = [filter_size * input.shape[1], num_filters]
......@@ -1740,6 +1741,7 @@ def sequence_last_step(input):
return sequence_pool(input=input, pool_type="last")
@templatedoc()
def pool2d(input,
pool_size=-1,
pool_type="max",
......@@ -1751,24 +1753,45 @@ def pool2d(input,
use_mkldnn=False,
name=None):
"""
This function adds the operator for pooling in 2 dimensions, using the
pooling configurations mentioned in input parameters.
${comment}
Args:
input (Variable): ${input_comment}
pool_size (int): ${ksize_comment}
pool_type (str): ${pooling_type_comment}
input (Variable): The input tensor of pooling operator. The format of
input tensor is NCHW, where N is batch size, C is
the number of channels, H is the height of the
feature, and W is the width of the feature.
pool_size (int): The side length of pooling windows. All pooling
windows are squares with pool_size on a side.
pool_type: ${pooling_type_comment}
pool_stride (int): stride of the pooling layer.
pool_padding (int): padding size.
global_pooling (bool): ${global_pooling_comment}
use_cudnn (bool): ${use_cudnn_comment}
ceil_mode (bool): ${ceil_mode_comment}
use_mkldnn (bool): ${use_mkldnn_comment}
name (str): A name for this layer(optional). If set None, the layer
will be named automatically.
global_pooling: ${global_pooling_comment}
use_cudnn: ${use_cudnn_comment}
ceil_mode: ${ceil_mode_comment}
use_mkldnn: ${use_mkldnn_comment}
name (str|None): A name for this layer(optional). If set None, the
layer will be named automatically.
Returns:
Variable: output of pool2d layer.
Variable: The pooling result.
Raises:
ValueError: If 'pool_type' is not "max" nor "avg"
ValueError: If 'global_pooling' is False and 'pool_size' is -1
ValueError: If 'use_cudnn' is not a bool value.
Examples:
.. code-block:: python
data = fluid.layers.data(
name='data', shape=[3, 32, 32], dtype='float32')
conv2d = fluid.layers.pool2d(
input=data,
pool_size=2,
pool_type='max',
pool_stride=1,
global_pooling=False)
"""
if pool_type not in ["max", "avg"]:
raise ValueError(
......@@ -2127,15 +2150,37 @@ def layer_norm(input,
def beam_search_decode(ids, scores, name=None):
"""
${beam_search_decode}
Beam Search Decode
This layers is to pack the output of beam search layer into sentences and
associated scores. It is usually called after the beam search layer.
Typically, the output of beam search layer is a tensor of selected ids, with
a tensor of the score of each id. Beam search layer's output ids, however,
are generated directly during the tree search, and they are stacked by each
level of the search tree. Thus we need to reorganize them into sentences,
based on the score of each id. This layer takes the output of beam search
layer as input and repack them into sentences.
Args:
ids (Variable): ${ids_comment}
scores (Variable): ${scores_comment}
ids (Variable): The selected ids, output of beam search layer.
scores (Variable): The associated scores of the ids, out put of beam
search layer.
name (str): The name of this layer. It is optional.
Returns:
tuple: a tuple of two output variable: sentence_ids, sentence_scores
tuple(Variable): a tuple of two output tensors: sentence_ids, sentence_scores.
sentence_ids is a tensor with shape [size, length], where size is the
beam size of beam search, and length is the length of each sentence.
Note that the length of sentences may vary.
sentence_scores is a tensor with the same shape as sentence_ids.
Examples:
.. code-block:: python
ids, scores = fluid.layers.beam_search(
pre_ids, ids, scores, beam_size, end_id)
sentence_ids, sentence_scores = fluid.layers.beam_search_decode(
ids, scores)
"""
helper = LayerHelper('beam_search_decode', **locals())
sentence_ids = helper.create_tmp_variable(dtype=ids.dtype)
......@@ -2567,7 +2612,7 @@ def beam_search(pre_ids, ids, scores, beam_size, end_id, level=0):
beam_size (int): ${beam_size_comment}
end_id (int): ${end_id_comment}
level (int): ${level_comment}
Returns:
tuple: a tuple of beam_search output variables: selected_ids, selected_scores
'''
......@@ -3016,7 +3061,7 @@ def split(input, num_or_sections, dim=-1, name=None):
will be named automatically.
Returns:
List: The list of segmented tensor variables.
list(Variable): The list of segmented tensor variables.
Examples:
.. code-block:: python
......@@ -3225,25 +3270,51 @@ def topk(input, k, name=None):
This operator is used to find values and indices of the k largest entries
for the last dimension.
If the input is a vector (rank=1), finds the k largest entries in the vector
If the input is a vector (1-D Tensor), finds the k largest entries in the vector
and outputs their values and indices as vectors. Thus values[j] is the j-th
largest entry in input, and its index is indices[j].
If the input is a Tensor with higher rank, this operator computes the top k
entries along the last dimension.
For example:
.. code-block:: text
If:
input = [[5, 4, 2, 3],
[9, 7, 10, 25],
[6, 2, 10, 1]]
k = 2
Then:
The first output:
values = [[5, 4],
[10, 25],
[6, 10]]
The second output:
indices = [[0, 1],
[2, 3],
[0, 2]]
Args:
input(Variable): The input variable which can be a vector or Tensor with
higher rank.
k(int): An integer value to specify the top k largest elements.
k(int): The number of top elements to look for along the last dimension
of input.
name(str|None): A name for this layer(optional). If set None, the layer
will be named automatically.
will be named automatically.
Default: None
Returns:
values(Variable): The k largest elements along each last dimensional
slice.
indices(Variable): The indices of values within the last dimension of
input.
Tuple[Variable]: A tuple with two elements. Each element is a Variable.
The first one is k largest elements along each last
dimensional slice. The second one is indices of values
within the last dimension of input.
Raises:
ValueError: If k < 1 or k is not less than the last dimension of input
Examples:
.. code-block:: python
......@@ -3251,7 +3322,7 @@ def topk(input, k, name=None):
top5_values, top5_indices = layers.topk(input, k=5)
"""
shape = input.shape
if k < 1 and k >= shape[-1]:
if k < 1 or k >= shape[-1]:
raise ValueError("k must be greater than 0 and less than %d." %
(shape[-1]))
......@@ -3269,8 +3340,7 @@ def topk(input, k, name=None):
return values, indices
def edit_distance(input, label, normalized=True, ignored_tokens=None,
name=None):
def edit_distance(input, label, normalized=True, ignored_tokens=None):
"""
EditDistance operator computes the edit distances between a batch of
hypothesis strings and their references. Edit distance, also called
......@@ -3284,21 +3354,21 @@ def edit_distance(input, label, normalized=True, ignored_tokens=None,
"kitten" -> "sitten" -> "sittin" -> "sitting"
Input(Hyps) is a LoDTensor consisting of all the hypothesis strings with
The input is a LoDTensor consisting of all the hypothesis strings with
the total number denoted by `batch_size`, and the separation is specified
by the LoD information. And the `batch_size` reference strings are arranged
in order in the same way in the LoDTensor Input(Refs).
in order in the same way in the input LoDTensor.
Output(Out) contains the `batch_size` results and each stands for the edit
The output contains the `batch_size` results and each stands for the edit
distance for a pair of strings respectively. If Attr(normalized) is true,
the edit distance will be divided by the length of reference string.
Args:
input(Variable): The indices for hypothesis strings.
label(Variable): The indices for reference strings.
normalized(bool): Indicated whether to normalize the edit distance by
normalized(bool, default True): Indicated whether to normalize the edit distance by
the length of reference string.
ignored_tokens(list of int): Tokens that should be removed before
ignored_tokens(list<int>, default None): Tokens that should be removed before
calculating edit distance.
name (str): The name of this layer. It is optional.
......@@ -3310,7 +3380,6 @@ def edit_distance(input, label, normalized=True, ignored_tokens=None,
x = fluid.layers.data(name='x', shape=[8], dtype='float32')
y = fluid.layers.data(name='y', shape=[7], dtype='float32')
cost = fluid.layers.edit_distance(input=x,label=y)
"""
helper = LayerHelper("edit_distance", **locals())
......@@ -3430,35 +3499,33 @@ def warpctc(input, label, blank=0, norm_by_times=False):
input tensor.
Args:
input(Variable): (LodTensor, default: LoDTensor<float>),
the unscaled probabilities of variable-length sequences,
which is a 2-D Tensor with LoD information.
It's shape is [Lp, num_classes + 1], where Lp is the sum of all input
sequences' length and num_classes is the true number of classes.
(not including the blank label).
label(Variable): (LodTensor, default: LoDTensor<int>), the ground truth
of variable-length sequence, which is a 2-D Tensor with LoD
information. It is of the shape [Lg, 1], where Lg is th sum of
all labels' length.
blank (int): default 0, the blank label index of Connectionist
Temporal Classification (CTC) loss, which is in the
half-opened interval [0, num_classes + 1).
norm_by_times (bool): default false, whether to normalize
the gradients by the number of time-step, which is also the
sequence's length. There is no need to normalize the gradients
if warpctc layer was follewed by a mean_op.
input (Variable): The unscaled probabilities of variable-length sequences,
which is a 2-D Tensor with LoD information.
It's shape is [Lp, num_classes + 1], where Lp is the sum of all input
sequences' length and num_classes is the true number of classes.
(not including the blank label).
label (Variable): The ground truth of variable-length sequence,
which is a 2-D Tensor with LoD information. It is of the shape [Lg, 1],
where Lg is th sum of all labels' length.
blank (int, default 0): The blank label index of Connectionist
Temporal Classification (CTC) loss, which is in the
half-opened interval [0, num_classes + 1).
norm_by_times(bool, default false): Whether to normalize the gradients
by the number of time-step, which is also the sequence's length.
There is no need to normalize the gradients if warpctc layer was
follewed by a mean_op.
Returns:
Variable: The Connectionist Temporal Classification (CTC) loss,
which is a 2-D Tensor of the shape [batch_size, 1].
Examples:
.. code-block:: python
y = layers.data(
name='y', shape=[11, 8], dtype='float32', lod_level=1)
y_predict = layers.data(
name='y_predict', shape=[11, 1], dtype='float32')
cost = layers.warpctc(input=y_predict, label=y)
label = fluid.layers.data(shape=[11, 8], dtype='float32', lod_level=1)
predict = fluid.layers.data(shape=[11, 1], dtype='float32')
cost = fluid.layers.warpctc(input=predict, label=label)
"""
helper = LayerHelper('warpctc', **locals())
......@@ -3487,17 +3554,21 @@ def sequence_reshape(input, new_dim):
.. code-block:: text
x is a LoDTensor:
x.lod = [[2, 4]]
x.data = [[1, 2], [3, 4],
[5, 6], [7, 8], [9, 10], [11, 12]]
x.lod = [[0, 2, 6]]
x.data = [[1, 2], [3, 4],
[5, 6], [7, 8],
[9, 10], [11, 12]]
x.dims = [6, 2]
set new_dim = 4
then out is a LoDTensor:
out.lod = [[1, 2]]
out.data = [[1, 2, 3, 4],
[5, 6, 7, 8], [9, 10, 11, 12]]
out.lod = [[0, 1, 3]]
out.data = [[1, 2, 3, 4],
[5, 6, 7, 8],
[9, 10, 11, 12]]
out.dims = [3, 4]
Currently, only 1-level LoDTensor is supported and please make sure
......@@ -3505,19 +3576,19 @@ def sequence_reshape(input, new_dim):
no remainder for each sequence.
Args:
input (Variable): (LodTensor, default: LoDTensor<float>), a 2-D LoDTensor
with shape being [N, M] where M for dimension.
new_dim (int): New dimension which the input LoDTensor is reshaped to.
input (Variable): A 2-D LoDTensor with shape being [N, M] where M for dimension.
new_dim (int): New dimension that the input LoDTensor is reshaped to.
Returns:
Variable: Reshaped LoDTensor according to new dimension.
Examples:
.. code-block:: python
x = fluid.layers.data(name='x', shape=[5, 20],
dtype='float32', lod_level=1)
x_reshaped = layers.sequence_reshape(input=x, new_dim=10)
x = fluid.layers.data(shape=[5, 20], dtype='float32', lod_level=1)
x_reshaped = fluid.layers.sequence_reshape(input=x, new_dim=10)
"""
helper = LayerHelper('sequence_reshape', **locals())
out = helper.create_tmp_variable(helper.input_dtype())
......@@ -3553,7 +3624,7 @@ def nce(input,
param_attr (ParamAttr|None): attributes for parameter
bias_attr (ParamAttr|None): attributes for bias
num_neg_samples (int): ${num_neg_samples_comment}
Returns:
Variable: The output nce loss.
......@@ -3723,8 +3794,6 @@ def im2sequence(input, filter_size=1, stride=1, padding=0, name=None):
Examples:
As an example:
.. code-block:: text
Given:
......@@ -3768,7 +3837,7 @@ def im2sequence(input, filter_size=1, stride=1, padding=0, name=None):
output.lod = [[4, 4]]
The simple usage is:
Examples:
.. code-block:: python
......@@ -4253,9 +4322,7 @@ def lrn(input, n=5, k=1.0, alpha=1e-4, beta=0.75, name=None):
.. math::
Output(i, x, y) = Input(i, x, y) / \left(
k + \alpha \sum\limits^{\min(C, c + n/2)}_{j = \max(0, c - n/2)}
(Input(j, x, y))^2 \right)^{\beta}
Output(i, x, y) = Input(i, x, y) / \\left(k + \\alpha \\sum\\limits^{\\min(C, c + n/2)}_{j = \\max(0, c - n/2)}(Input(j, x, y))^2\\right)^{\\beta}
In the above equation:
......@@ -4769,6 +4836,62 @@ def random_crop(x, shape, seed=None):
return out
def log(x):
"""
Calculates the natural log of the given input tensor, element-wise.
.. math::
Out = \\ln(x)
Args:
x (Variable): Input tensor.
Returns:
Variable: The natural log of the input tensor computed element-wise.
Examples:
.. code-block:: python
output = fluid.layers.log(x)
"""
helper = LayerHelper('log', **locals())
dtype = helper.input_dtype()
out = helper.create_tmp_variable(dtype)
helper.append_op(type="log", inputs={"X": input}, outputs={"Out": out})
return out
def relu(x):
"""
Relu takes one input data (Tensor) and produces one output data (Tensor)
where the rectified linear function, y = max(0, x), is applied to
the tensor elementwise.
.. math::
Out = \\max(0, x)
Args:
x (Variable): The input tensor.
Returns:
Variable: The output tensor with the same shape as input.
Examples:
.. code-block:: python
output = fluid.layers.relu(x)
"""
helper = LayerHelper('relu', **locals())
dtype = helper.input_dtype()
out = helper.create_tmp_variable(dtype)
helper.append_op(type="relu", inputs={"X": input}, outputs={"Out": out})
return out
def mean_iou(input, label, num_classes):
"""
Mean Intersection-Over-Union is a common evaluation metric for
......@@ -4795,11 +4918,10 @@ def mean_iou(input, label, num_classes):
out_wrong(Variable): A Tensor with shape [num_classes]. The wrong numbers of each class.
out_correct(Variable): A Tensor with shape [num_classes]. The correct numbers of each class.
Examples:
.. code-block:: python
iou, wrongs, corrects = fluid.layers.mean_iou(predict, label, num_classes)
"""
helper = LayerHelper('mean_iou', **locals())
......
python/paddle/fluid/layers/ops.py
浏览文件 @ b77c886e
......@@ -17,7 +17,6 @@ __activations__ = [
'sigmoid',
'logsigmoid',
'exp',
'relu',
'tanh',
'tanh_shrink',
'softshrink',
......@@ -29,7 +28,6 @@ __activations__ = [
'sin',
'round',
'reciprocal',
'log',
'square',
'softplus',
'softsign',
......
python/paddle/fluid/layers/tensor.py
浏览文件 @ b77c886e
......@@ -108,16 +108,29 @@ def create_global_var(shape,
force_cpu=False,
name=None):
"""
Create a global variable. such as global_step
Create a new variable in the global block(block 0).
Args:
shape(list[int]): shape of the variable
value(float): the value of the variable
dtype(string): element type of the parameter
persistable(bool): if this variable is persistable
force_cpu(bool): force this variable to be on CPU
value(float): the value of the variable. The new created
variable will be filled with it.
dtype(string): data type of the variable
persistable(bool): if this variable is persistable.
Default: False
force_cpu(bool): force this variable to be on CPU.
Default: False
name(str|None): The name of the variable. If set to None the variable
name will be generated automatically.
Default: None
Returns:
Variable: the created Variable
Examples:
.. code-block:: python
var = fluid.create_global_var(shape=[2,3], value=1.0, dtype='float32',
persistable=True, force_cpu=True, name='new_var')
"""
helper = LayerHelper("global_var", **locals())
var = helper.create_global_variable(
......@@ -175,7 +188,8 @@ def concat(input, axis=0, name=None):
Examples:
.. code-block:: python
out = fluid.layers.concat(input=[Efirst, Esecond, Ethird, Efourth])
out = fluid.layers.concat(input=[Efirst, Esecond, Ethird, Efourth])
"""
helper = LayerHelper('concat', **locals())
out = helper.create_tmp_variable(dtype=helper.input_dtype())
......@@ -188,19 +202,21 @@ def concat(input, axis=0, name=None):
def sums(input, out=None):
"""This function performs the sum operation on the input and returns the
"""
This function performs the sum operation on the input and returns the
result as the output.
Args:
input (Variable|list): The input tensor that has the elements
that need to be summed up.
out (Variable|None): Output parameter. The sum result.
Default: None
Returns:
Variable: The tensor type variable that has the sum of input
written to it.
Variable: the sum of input. The same as the argument 'out'
Examples:
.. code-block::python
.. code-block:: python
tmp = fluid.layers.zeros(shape=[10], dtype='int32')
i = fluid.layers.fill_constant(shape=[1], dtype='int64', value=10)
......@@ -371,13 +387,13 @@ def argmin(x, axis=0):
x(Variable): The input to compute the indices of
the min elements.
axis(int): Axis to compute indices along.
Returns:
Variable: The tensor variable storing the output
Examples:
.. code-block:: python
out = fluid.layers.argmin(x=in, axis=0)
out = fluid.layers.argmin(x=in, axis=-1)
"""
......@@ -402,13 +418,13 @@ def argmax(x, axis=0):
x(Variable): The input to compute the indices of
the max elements.
axis(int): Axis to compute indices along.
Returns:
Variable: The tensor variable storing the output
Examples:
.. code-block:: python
out = fluid.layers.argmax(x=in, axis=0)
out = fluid.layers.argmax(x=in, axis=-1)
"""
......@@ -456,11 +472,12 @@ def zeros(shape, dtype, force_cpu=False):
It also sets *stop_gradient* to True.
Args:
shape(tuple|list|None): Shape of output tensor
dtype(np.dtype|core.VarDesc.VarType|str): Data type of output tensor
shape(tuple|list|None): Shape of output tensor.
dtype(np.dtype|core.VarDesc.VarType|str): Data type of output tensor.
force_cpu(bool, default False): Whether to make output stay on CPU.
Returns:
Variable: The tensor variable storing the output
Variable: The tensor variable storing the output.
Examples:
.. code-block:: python
......
Markdown is supported
0% 或 .
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 或