mindspore-Ecosystem
/
mindspore
21
0
Fork 27
Code Issues Pull Requests 6 Projects Releases Wiki Activity

fix docs error of probability

This commit is contained in:
qianjiahong 2021-07-09 11:44:40 +08:00
parent 4bcf62b171
commit 6f9b1b656f
18 changed files with 88 additions and 56 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
mindspore/nn/probability/bijector/bijector.py
View File

@ -278,7 +278,7 @@ class Bijector(Cell):
Args:
name (str): The name of the function.
*args (list): A list of positional arguments that the function needs.
**kwargs (dictionary): A dictionary of keyword arguments that the function needs.
**kwargs (dict): A dictionary of keyword arguments that the function needs.
"""
if name == 'forward':
return self.forward(*args, **kwargs)

3
mindspore/nn/probability/bijector/invert.py
View File

@ -29,6 +29,7 @@ class Invert(Bijector):
``Ascend`` ``GPU``
Examples:
>>> import numpy as np
>>> import mindspore
>>> import mindspore.nn as nn
>>> import mindspore.nn.probability.bijector as msb
@ -44,6 +45,8 @@ class Invert(Bijector):
>>> forward = Net()
>>> x = np.array([2.0, 3.0, 4.0, 5.0]).astype(np.float32)
>>> ans = forward(Tensor(x, dtype=mindspore.float32))
>>> print(ans.shape)
(4,)
"""
def __init__(self,

17
mindspore/nn/probability/bnn_layers/bnn_cell_wrapper.py
View File

@ -41,15 +41,26 @@ class WithBNNLossCell(Cell):
``Ascend`` ``GPU``
Examples:
>>> import numpy as np
>>> import mindspore.nn as nn
>>> from mindspore.nn.probability import bnn_layers
>>> from mindspore import Tensor
>>> class Net(nn.Cell):
... def __init__(self):
... super(Net, self).__init__()
... self.dense = bnn_layers.DenseReparam(16, 1)
... def construct(self, x):
... return self.dense(x)
>>> net = Net()
>>> loss_fn = nn.SoftmaxCrossEntropyWithLogits(sparse=False)
>>> net_with_criterion = WithBNNLossCell(net, loss_fn)
>>> net_with_criterion = bnn_layers.WithBNNLossCell(net, loss_fn)
>>>
>>> batch_size = 2
>>> data = Tensor(np.ones([batch_size, 16]).astype(np.float32) * 0.01)
>>> label = Tensor(np.ones([batch_size, 1]).astype(np.float32))
>>>
>>> net_with_criterion(data, label)
>>> output = net_with_criterion(data, label)
>>> print(output.shape)
(2,)
"""
def __init__(self, backbone, loss_fn, dnn_factor=1, bnn_factor=1):

14
mindspore/nn/probability/bnn_layers/conv_variational.py
View File

@ -191,19 +191,19 @@ class ConvReparam(_ConvVariational):
Default: 1.
has_bias (bool): Specifies whether the layer uses a bias vector.
Default: False.
weight_prior_fn: The prior distribution for weight.
weight_prior_fn (Cell): The prior distribution for weight.
It must return a mindspore distribution instance.
Default: NormalPrior. (which creates an instance of standard
normal distribution). The current version only supports normal distribution.
weight_posterior_fn: The posterior distribution for sampling weight.
weight_posterior_fn (function): The posterior distribution for sampling weight.
It must be a function handle which returns a mindspore
distribution instance. Default: normal_post_fn.
The current version only supports normal distribution.
bias_prior_fn: The prior distribution for bias vector. It must return
bias_prior_fn (Cell): The prior distribution for bias vector. It must return
a mindspore distribution. Default: NormalPrior(which creates an
instance of standard normal distribution). The current version
only supports normal distribution.
bias_posterior_fn: The posterior distribution for sampling bias vector.
bias_posterior_fn (function): The posterior distribution for sampling bias vector.
It must be a function handle which returns a mindspore
distribution instance. Default: normal_post_fn.
The current version only supports normal distribution.
@ -218,7 +218,11 @@ class ConvReparam(_ConvVariational):
``Ascend`` ``GPU``
Examples:
>>> net = ConvReparam(120, 240, 4, has_bias=False)
>>> import numpy as np
>>> import mindspore
>>> from mindspore import Tensor
>>> from mindspore.nn.probability import bnn_layers
>>> net = bnn_layers.ConvReparam(120, 240, 4, has_bias=False)
>>> input = Tensor(np.ones([1, 120, 1024, 640]), mindspore.float32)
>>> output = net(input).shape
>>> print(output)

28
mindspore/nn/probability/bnn_layers/dense_variational.py
View File

@ -136,19 +136,19 @@ class DenseReparam(_DenseVariational):
Note that if the type of activation is Cell, it must be instantiated beforehand.
Default: None.
has_bias (bool): Specifies whether the layer uses a bias vector. Default: False.
weight_prior_fn: The prior distribution for weight.
weight_prior_fn (Cell): The prior distribution for weight.
It must return a mindspore distribution instance.
Default: NormalPrior. (which creates an instance of standard
normal distribution). The current version only supports normal distribution.
weight_posterior_fn: The posterior distribution for sampling weight.
weight_posterior_fn (function): The posterior distribution for sampling weight.
It must be a function handle which returns a mindspore
distribution instance. Default: normal_post_fn.
The current version only supports normal distribution.
bias_prior_fn: The prior distribution for bias vector. It must return
bias_prior_fn (Cell): The prior distribution for bias vector. It must return
a mindspore distribution. Default: NormalPrior(which creates an
instance of standard normal distribution). The current version
only supports normal distribution.
bias_posterior_fn: The posterior distribution for sampling bias vector.
bias_posterior_fn (function): The posterior distribution for sampling bias vector.
It must be a function handle which returns a mindspore
distribution instance. Default: normal_post_fn.
The current version only supports normal distribution.
@ -163,7 +163,11 @@ class DenseReparam(_DenseVariational):
``Ascend`` ``GPU``
Examples:
>>> net = DenseReparam(3, 4)
>>> import numpy as np
>>> import mindspore
>>> from mindspore import Tensor
>>> from mindspore.nn.probability import bnn_layers
>>> net = bnn_layers.DenseReparam(3, 4)
>>> input = Tensor(np.random.randint(0, 255, [2, 3]), mindspore.float32)
>>> output = net(input).shape
>>> print(output)
@ -225,19 +229,19 @@ class DenseLocalReparam(_DenseVariational):
Note that if the type of activation is Cell, it must be instantiated beforehand.
Default: None.
has_bias (bool): Specifies whether the layer uses a bias vector. Default: False.
weight_prior_fn: The prior distribution for weight.
weight_prior_fn (Cell): The prior distribution for weight.
It must return a mindspore distribution instance.
Default: NormalPrior. (which creates an instance of standard
normal distribution). The current version only supports normal distribution.
weight_posterior_fn: The posterior distribution for sampling weight.
weight_posterior_fn (function): The posterior distribution for sampling weight.
It must be a function handle which returns a mindspore
distribution instance. Default: normal_post_fn.
The current version only supports normal distribution.
bias_prior_fn: The prior distribution for bias vector. It must return
bias_prior_fn (Cell): The prior distribution for bias vector. It must return
a mindspore distribution. Default: NormalPrior(which creates an
instance of standard normal distribution). The current version
only supports normal distribution.
bias_posterior_fn: The posterior distribution for sampling bias vector.
bias_posterior_fn (function): The posterior distribution for sampling bias vector.
It must be a function handle which returns a mindspore
distribution instance. Default: normal_post_fn.
The current version only supports normal distribution.
@ -252,7 +256,11 @@ class DenseLocalReparam(_DenseVariational):
``Ascend`` ``GPU``
Examples:
>>> net = DenseLocalReparam(3, 4)
>>> import numpy as np
>>> import mindspore
>>> from mindspore import Tensor
>>> from mindspore.nn.probability import bnn_layers
>>> net = bnn_layers.DenseLocalReparam(3, 4)
>>> input = Tensor(np.random.randint(0, 255, [2, 3]), mindspore.float32)
>>> output = net(input).shape
>>> print(output)

4
mindspore/nn/probability/distribution/bernoulli.py
View File

@ -282,8 +282,8 @@ class Bernoulli(Distribution):
Args:
dist (str): The type of the distributions. Should be "Bernoulli" in this case.
probs1_b (Union[Tensor, numbers.Number]): `probs1` of distribution b.
probs1_a (Union[Tensor, numbers.Number]): `probs1` of distribution a. Default: self.probs.
probs1_b (Tensor, Number): `probs1` of distribution b.
probs1_a (Tensor, Number): `probs1` of distribution a. Default: self.probs.
.. math::
KL(a||b) = probs1_a * \log(\frac{probs1_a}{probs1_b}) +

32
mindspore/nn/probability/distribution/distribution.py
View File

@ -362,7 +362,7 @@ class Distribution(Cell):
Args:
*args (list): the list of positional arguments forwarded to subclasses.
**kwargs (dictionary): the dictionary of keyword arguments forwarded to subclasses.
**kwargs (dict): the dictionary of keyword arguments forwarded to subclasses.
Note:
`dist_spec_args` must be passed in through list or dictionary. The order of `dist_spec_args`
@ -394,7 +394,7 @@ class Distribution(Cell):
Args:
value (Tensor): value to be evaluated.
*args (list): the list of positional arguments forwarded to subclasses.
**kwargs (dictionary): the dictionary of keyword arguments forwarded to subclasses.
**kwargs (dict): the dictionary of keyword arguments forwarded to subclasses.
Note:
A distribution can be optionally passed to the function by passing its dist_spec_args through
@ -418,7 +418,7 @@ class Distribution(Cell):
Args:
value (Tensor): value to be evaluated.
*args (list): the list of positional arguments forwarded to subclasses.
**kwargs (dictionary): the dictionary of keyword arguments forwarded to subclasses.
**kwargs (dict): the dictionary of keyword arguments forwarded to subclasses.
Note:
A distribution can be optionally passed to the function by passing its dist_spec_args through
@ -442,7 +442,7 @@ class Distribution(Cell):
Args:
value (Tensor): value to be evaluated.
*args (list): the list of positional arguments forwarded to subclasses.
**kwargs (dictionary): the dictionary of keyword arguments forwarded to subclasses.
**kwargs (dict): the dictionary of keyword arguments forwarded to subclasses.
Note:
A distribution can be optionally passed to the function by passing its dist_spec_args through
@ -484,7 +484,7 @@ class Distribution(Cell):
Args:
value (Tensor): value to be evaluated.
*args (list): the list of positional arguments forwarded to subclasses.
**kwargs (dictionary): the dictionary of keyword arguments forwarded to subclasses.
**kwargs (dict: the dictionary of keyword arguments forwarded to subclasses.
Note:
A distribution can be optionally passed to the function by passing its dist_spec_args through
@ -508,7 +508,7 @@ class Distribution(Cell):
Args:
value (Tensor): value to be evaluated.
*args (list): the list of positional arguments forwarded to subclasses.
**kwargs (dictionary): the dictionary of keyword arguments forwarded to subclasses.
**kwargs (dict): the dictionary of keyword arguments forwarded to subclasses.
Note:
A distribution can be optionally passed to the function by passing its dist_spec_args through
@ -541,7 +541,7 @@ class Distribution(Cell):
Args:
value (Tensor): value to be evaluated.
*args (list): the list of positional arguments forwarded to subclasses.
**kwargs (dictionary): the dictionary of keyword arguments forwarded to subclasses.
**kwargs (dict): the dictionary of keyword arguments forwarded to subclasses.
Note:
A distribution can be optionally passed to the function by passing its dist_spec_args through
@ -568,7 +568,7 @@ class Distribution(Cell):
Args:
dist (str): type of the distribution.
*args (list): the list of positional arguments forwarded to subclasses.
**kwargs (dictionary): the dictionary of keyword arguments forwarded to subclasses.
**kwargs (dict): the dictionary of keyword arguments forwarded to subclasses.
Note:
dist_spec_args of distribution b must be passed to the function through `args` or `kwargs`.
@ -585,7 +585,7 @@ class Distribution(Cell):
Args:
*args (list): the list of positional arguments forwarded to subclasses.
**kwargs (dictionary): the dictionary of keyword arguments forwarded to subclasses.
**kwargs (dict): the dictionary of keyword arguments forwarded to subclasses.
Note:
A distribution can be optionally passed to the function by passing its *dist_spec_args* through
@ -602,7 +602,7 @@ class Distribution(Cell):
Args:
*args (list): the list of positional arguments forwarded to subclasses.
**kwargs (dictionary): the dictionary of keyword arguments forwarded to subclasses.
**kwargs (dict): the dictionary of keyword arguments forwarded to subclasses.
Note:
A distribution can be optionally passed to the function by passing its *dist_spec_args* through
@ -616,7 +616,7 @@ class Distribution(Cell):
Args:
*args (list): the list of positional arguments forwarded to subclasses.
**kwargs (dictionary): the dictionary of keyword arguments forwarded to subclasses.
**kwargs (dict: the dictionary of keyword arguments forwarded to subclasses.
Note:
A distribution can be optionally passed to the function by passing its *dist_spec_args* through
@ -630,7 +630,7 @@ class Distribution(Cell):
Args:
*args (list): the list of positional arguments forwarded to subclasses.
**kwargs (dictionary): the dictionary of keyword arguments forwarded to subclasses.
**kwargs (dict): the dictionary of keyword arguments forwarded to subclasses.
Note:
A distribution can be optionally passed to the function by passing its *dist_spec_args* through
@ -665,7 +665,7 @@ class Distribution(Cell):
Args:
*args (list): the list of positional arguments forwarded to subclasses.
**kwargs (dictionary): the dictionary of keyword arguments forwarded to subclasses.
**kwargs (dict): the dictionary of keyword arguments forwarded to subclasses.
Note:
A distribution can be optionally passed to the function by passing its *dist_spec_args* through
@ -680,7 +680,7 @@ class Distribution(Cell):
Args:
dist (str): type of the distribution.
*args (list): the list of positional arguments forwarded to subclasses.
**kwargs (dictionary): the dictionary of keyword arguments forwarded to subclasses.
**kwargs (dict): the dictionary of keyword arguments forwarded to subclasses.
Note:
dist_spec_args of distribution b must be passed to the function through `args` or `kwargs`.
@ -707,7 +707,7 @@ class Distribution(Cell):
Args:
shape (tuple): shape of the sample.
*args (list): the list of positional arguments forwarded to subclasses.
**kwargs (dictionary): the dictionary of keyword arguments forwarded to subclasses.
**kwargs (dict): the dictionary of keyword arguments forwarded to subclasses.
Note:
A distribution can be optionally passed to the function by passing its *dist_spec_args* through
@ -728,7 +728,7 @@ class Distribution(Cell):
Args:
name (str): The name of the function.
*args (list): A list of positional arguments that the function needs.
**kwargs (dictionary): A dictionary of keyword arguments that the function needs.
**kwargs (dict): A dictionary of keyword arguments that the function needs.
"""
if name == 'log_prob':

4
mindspore/nn/probability/distribution/gamma.py
View File

@ -30,9 +30,9 @@ class Gamma(Distribution):
Args:
concentration (list, numpy.ndarray, Tensor): The concentration,
also know as alpha of the Gamma distribution.
also know as alpha of the Gamma distribution. Default: None.
rate (list, numpy.ndarray, Tensor): The rate, also know as
beta of the Gamma distribution.
beta of the Gamma distribution. Default: None.
seed (int): The seed used in sampling. The global seed is used if it is None. Default: None.
dtype (mindspore.dtype): The type of the event samples. Default: mstype.float32.
name (str): The name of the distribution. Default: 'Gamma'.

2
mindspore/nn/probability/distribution/geometric.py
View File

@ -31,7 +31,7 @@ class Geometric(Distribution):
trials when the first success is achieved.
Args:
probs (float, list, numpy.ndarray, Tensor): The probability of success.
probs (float, list, numpy.ndarray, Tensor): The probability of success. Default: None.
seed (int): The seed used in sampling. Global seed is used if it is None. Default: None.
dtype (mindspore.dtype): The type of the event samples. Default: mstype.int32.
name (str): The name of the distribution. Default: 'Geometric'.

7
mindspore/nn/probability/distribution/log_normal.py
View File

@ -29,9 +29,9 @@ class LogNormal(msd.TransformedDistribution):
logarithm is normally distributed. It is constructed as the exponential transformation of a Normal distribution.
Args:
loc (int, float, list, numpy.ndarray, Tensor): The mean of the underlying Normal distribution.
loc (int, float, list, numpy.ndarray, Tensor): The mean of the underlying Normal distribution. Default: None.
scale (int, float, list, numpy.ndarray, Tensor): The standard deviation of the underlying
Normal distribution.
Normal distribution. Default: None.
seed (int): the seed used in sampling. The global seed is used if it is None. Default: None.
dtype (mindspore.dtype): type of the distribution. Default: mstype.float32.
name (str): the name of the distribution. Default: 'LogNormal'.
@ -45,6 +45,7 @@ class LogNormal(msd.TransformedDistribution):
`dtype` must be a float type because LogNormal distributions are continuous.
Examples:
>>> import numpy as np
>>> import mindspore
>>> import mindspore.nn as nn
>>> import mindspore.nn.probability.distribution as msd
@ -57,6 +58,8 @@ class LogNormal(msd.TransformedDistribution):
... return self.ln.prob(x_)
>>> pdf = Prob()
>>> output = pdf(Tensor([1.0, 2.0], dtype=mindspore.float32))
>>> print(output.shape)
(2, 2)
"""
def __init__(self,

4
mindspore/nn/probability/distribution/logistic.py
View File

@ -28,8 +28,8 @@ class Logistic(Distribution):
Logistic distribution.
Args:
loc (int, float, list, numpy.ndarray, Tensor): The location of the Logistic distribution.
scale (int, float, list, numpy.ndarray, Tensor): The scale of the Logistic distribution.
loc (int, float, list, numpy.ndarray, Tensor): The location of the Logistic distribution. Default: None.
scale (int, float, list, numpy.ndarray, Tensor): The scale of the Logistic distribution. Default: None.
seed (int): The seed used in sampling. The global seed is used if it is None. Default: None.
dtype (mindspore.dtype): The type of the event samples. Default: mstype.float32.
name (str): The name of the distribution. Default: 'Logistic'.

4
mindspore/nn/probability/distribution/normal.py
View File

@ -28,8 +28,8 @@ class Normal(Distribution):
Normal distribution.
Args:
mean (int, float, list, numpy.ndarray, Tensor): The mean of the Normal distribution.
sd (int, float, list, numpy.ndarray, Tensor): The standard deviation of the Normal distribution.
mean (int, float, list, numpy.ndarray, Tensor): The mean of the Normal distribution. Default: None.
sd (int, float, list, numpy.ndarray, Tensor): The standard deviation of the Normal distribution. Default: None.
seed (int): The seed used in sampling. The global seed is used if it is None. Default: None.
dtype (mindspore.dtype): The type of the event samples. Default: mstype.float32.
name (str): The name of the distribution. Default: 'Normal'.

2
mindspore/nn/probability/distribution/poisson.py
View File

@ -29,7 +29,7 @@ class Poisson(Distribution):
Poisson Distribution.
Args:
rate (list, numpy.ndarray, Tensor): The rate of the Poisson distribution..
rate (list, numpy.ndarray, Tensor): The rate of the Poisson distribution. Default: None.
seed (int): The seed used in sampling. The global seed is used if it is None. Default: None.
dtype (mindspore.dtype): The type of the event samples. Default: mstype.float32.
name (str): The name of the distribution. Default: 'Poisson'.

3
mindspore/nn/probability/distribution/transformed_distribution.py
View File

@ -52,6 +52,7 @@ class TransformedDistribution(Distribution):
`reset_parameters` followed by `add_parameter`.
Examples:
>>> import numpy as np
>>> import mindspore
>>> import mindspore.nn as nn
>>> import mindspore.nn.probability.distribution as msd
@ -75,6 +76,8 @@ class TransformedDistribution(Distribution):
>>> x = np.array([2.0, 3.0, 4.0, 5.0]).astype(np.float32)
>>> tx = Tensor(x, dtype=mindspore.float32)
>>> cdf, sample = net(tx)
>>> print(sample.shape)
(2, 3)
"""
def __init__(self,

4
mindspore/nn/probability/distribution/uniform.py
View File

@ -28,8 +28,8 @@ class Uniform(Distribution):
Example class: Uniform Distribution.
Args:
low (int, float, list, numpy.ndarray, Tensor): The lower bound of the distribution.
high (int, float, list, numpy.ndarray, Tensor): The upper bound of the distribution.
low (int, float, list, numpy.ndarray, Tensor): The lower bound of the distribution. Default: None.
high (int, float, list, numpy.ndarray, Tensor): The upper bound of the distribution. Default: None.
seed (int): The seed uses in sampling. The global seed is used if it is None. Default: None.
dtype (mindspore.dtype): The type of the event samples. Default: mstype.float32.
name (str): The name of the distribution. Default: 'Uniform'.

4
mindspore/nn/probability/toolbox/anomaly_detection.py
View File

@ -33,8 +33,8 @@ class VAEAnomalyDetection:
Args:
encoder(Cell): The Deep Neural Network (DNN) model defined as encoder.
decoder(Cell): The DNN model defined as decoder.
hidden_size(int): The size of encoder's output tensor.
latent_size(int): The size of the latent space.
hidden_size(int): The size of encoder's output tensor. Default: 400.
latent_size(int): The size of the latent space. Default: 20.
Supported Platforms:
``Ascend`` ``GPU``

6
mindspore/nn/probability/toolbox/uncertainty_evaluation.py
View File

@ -58,10 +58,8 @@ class UncertaintyEvaluation:
Examples:
>>> network = LeNet()
>>> param_dict = load_checkpoint('checkpoint_lenet.ckpt')
>>> load_param_into_net(network, param_dict)
>>> ds_train = create_dataset('workspace/mnist/train')
>>> ds_eval = create_dataset('workspace/mnist/test')
>>> ds_train = create_dataset('workspace/mnist/train') # handle train data
>>> ds_eval = create_dataset('workspace/mnist/test') # handle test data
>>> evaluation = UncertaintyEvaluation(model=network,
... train_dataset=ds_train,
... task_type='classification',

4
mindspore/nn/probability/transforms/transform_bnn.py
View File

@ -161,7 +161,9 @@ class TransformToBNN:
``Ascend`` ``GPU``
Examples:
>>> net = Net()
>>> import mindspore.nn as nn
>>> from mindspore.nn.probability import bnn_layers
>>> net = LeNet()
>>> criterion = nn.SoftmaxCrossEntropyWithLogits(sparse=True)
>>> optim = nn.AdamWeightDecay(params=net.trainable_params(), learning_rate=0.0001)
>>> net_with_loss = nn.WithLossCell(net, criterion)