LostTech.TensorFlow : API Documentation

Type Bernoulli

Namespace tensorflow.distributions

Parent Distribution

Interfaces IBernoulli

Bernoulli distribution.

The Bernoulli distribution with `probs` parameter, i.e., the probability of a `1` outcome (vs a `0` outcome).

Methods

Properties

Public instance methods

object cdf(double value, string name)

Cumulative distribution function.

Given random variable `X`, the cumulative distribution function `cdf` is:

```none cdf(x) := P[X <= x] ```
Parameters
double value
`float` or `double` `Tensor`.
string name
Python `str` prepended to names of ops created by this function.
Returns
object

object cdf(ndarray value, string name)

Cumulative distribution function.

Given random variable `X`, the cumulative distribution function `cdf` is:

```none cdf(x) := P[X <= x] ```
Parameters
ndarray value
`float` or `double` `Tensor`.
string name
Python `str` prepended to names of ops created by this function.
Returns
object

object cdf(IEnumerable<double> value, string name)

Cumulative distribution function.

Given random variable `X`, the cumulative distribution function `cdf` is:

```none cdf(x) := P[X <= x] ```
Parameters
IEnumerable<double> value
`float` or `double` `Tensor`.
string name
Python `str` prepended to names of ops created by this function.
Returns
object

object cdf(IGraphNodeBase value, string name)

Cumulative distribution function.

Given random variable `X`, the cumulative distribution function `cdf` is:

```none cdf(x) := P[X <= x] ```
Parameters
IGraphNodeBase value
`float` or `double` `Tensor`.
string name
Python `str` prepended to names of ops created by this function.
Returns
object

object cdf(int value, string name)

Cumulative distribution function.

Given random variable `X`, the cumulative distribution function `cdf` is:

```none cdf(x) := P[X <= x] ```
Parameters
int value
`float` or `double` `Tensor`.
string name
Python `str` prepended to names of ops created by this function.
Returns
object

object cdf(IndexedSlices value, string name)

Cumulative distribution function.

Given random variable `X`, the cumulative distribution function `cdf` is:

```none cdf(x) := P[X <= x] ```
Parameters
IndexedSlices value
`float` or `double` `Tensor`.
string name
Python `str` prepended to names of ops created by this function.
Returns
object

object cdf_dyn(object value, ImplicitContainer<T> name)

Cumulative distribution function.

Given random variable `X`, the cumulative distribution function `cdf` is:

```none cdf(x) := P[X <= x] ```
Parameters
object value
`float` or `double` `Tensor`.
ImplicitContainer<T> name
Python `str` prepended to names of ops created by this function.
Returns
object

object copy(IDictionary<string, object> override_parameters_kwargs)

Creates a deep copy of the distribution.

Note: the copy distribution may continue to depend on the original initialization arguments.
Parameters
IDictionary<string, object> override_parameters_kwargs
String/value dictionary of initialization arguments to override with new values.
Returns
object

object copy_dyn(IDictionary<string, object> override_parameters_kwargs)

Creates a deep copy of the distribution.

Note: the copy distribution may continue to depend on the original initialization arguments.
Parameters
IDictionary<string, object> override_parameters_kwargs
String/value dictionary of initialization arguments to override with new values.
Returns
object

Tensor event_shape_tensor(string name)

Shape of a single sample from a single batch as a 1-D int32 `Tensor`.
Parameters
string name
name to give to the op
Returns
Tensor

object event_shape_tensor_dyn(ImplicitContainer<T> name)

Shape of a single sample from a single batch as a 1-D int32 `Tensor`.
Parameters
ImplicitContainer<T> name
name to give to the op
Returns
object

Tensor is_scalar_batch(string name)

Indicates that `batch_shape == []`.
Parameters
string name
Python `str` prepended to names of ops created by this function.
Returns
Tensor

object is_scalar_batch_dyn(ImplicitContainer<T> name)

Indicates that `batch_shape == []`.
Parameters
ImplicitContainer<T> name
Python `str` prepended to names of ops created by this function.
Returns
object

Tensor is_scalar_event(string name)

Indicates that `event_shape == []`.
Parameters
string name
Python `str` prepended to names of ops created by this function.
Returns
Tensor

object is_scalar_event_dyn(ImplicitContainer<T> name)

Indicates that `event_shape == []`.
Parameters
ImplicitContainer<T> name
Python `str` prepended to names of ops created by this function.
Returns
object

object log_cdf(IGraphNodeBase value, string name)

Log cumulative distribution function.

Given random variable `X`, the cumulative distribution function `cdf` is:

```none log_cdf(x) := Log[ P[X <= x] ] ```

Often, a numerical approximation can be used for `log_cdf(x)` that yields a more accurate answer than simply taking the logarithm of the `cdf` when `x << -1`.
Parameters
IGraphNodeBase value
`float` or `double` `Tensor`.
string name
Python `str` prepended to names of ops created by this function.
Returns
object

object log_cdf(IndexedSlices value, string name)

Log cumulative distribution function.

Given random variable `X`, the cumulative distribution function `cdf` is:

```none log_cdf(x) := Log[ P[X <= x] ] ```

Often, a numerical approximation can be used for `log_cdf(x)` that yields a more accurate answer than simply taking the logarithm of the `cdf` when `x << -1`.
Parameters
IndexedSlices value
`float` or `double` `Tensor`.
string name
Python `str` prepended to names of ops created by this function.
Returns
object

object log_cdf(IEnumerable<double> value, string name)

Log cumulative distribution function.

Given random variable `X`, the cumulative distribution function `cdf` is:

```none log_cdf(x) := Log[ P[X <= x] ] ```

Often, a numerical approximation can be used for `log_cdf(x)` that yields a more accurate answer than simply taking the logarithm of the `cdf` when `x << -1`.
Parameters
IEnumerable<double> value
`float` or `double` `Tensor`.
string name
Python `str` prepended to names of ops created by this function.
Returns
object

object log_cdf(ndarray value, string name)

Log cumulative distribution function.

Given random variable `X`, the cumulative distribution function `cdf` is:

```none log_cdf(x) := Log[ P[X <= x] ] ```

Often, a numerical approximation can be used for `log_cdf(x)` that yields a more accurate answer than simply taking the logarithm of the `cdf` when `x << -1`.
Parameters
ndarray value
`float` or `double` `Tensor`.
string name
Python `str` prepended to names of ops created by this function.
Returns
object

object log_cdf(double value, string name)

Log cumulative distribution function.

Given random variable `X`, the cumulative distribution function `cdf` is:

```none log_cdf(x) := Log[ P[X <= x] ] ```

Often, a numerical approximation can be used for `log_cdf(x)` that yields a more accurate answer than simply taking the logarithm of the `cdf` when `x << -1`.
Parameters
double value
`float` or `double` `Tensor`.
string name
Python `str` prepended to names of ops created by this function.
Returns
object

object log_cdf_dyn(object value, ImplicitContainer<T> name)

Log cumulative distribution function.

Given random variable `X`, the cumulative distribution function `cdf` is:

```none log_cdf(x) := Log[ P[X <= x] ] ```

Often, a numerical approximation can be used for `log_cdf(x)` that yields a more accurate answer than simply taking the logarithm of the `cdf` when `x << -1`.
Parameters
object value
`float` or `double` `Tensor`.
ImplicitContainer<T> name
Python `str` prepended to names of ops created by this function.
Returns
object

object log_prob(object value, string name)

Log probability density/mass function.
Parameters
object value
`float` or `double` `Tensor`.
string name
Python `str` prepended to names of ops created by this function.
Returns
object

object log_prob_dyn(object value, ImplicitContainer<T> name)

Log probability density/mass function.
Parameters
object value
`float` or `double` `Tensor`.
ImplicitContainer<T> name
Python `str` prepended to names of ops created by this function.
Returns
object

Tensor log_survival_function(ndarray value, string name)

Log survival function.

Given random variable `X`, the survival function is defined:

```none log_survival_function(x) = Log[ P[X > x] ] = Log[ 1 - P[X <= x] ] = Log[ 1 - cdf(x) ] ```

Typically, different numerical approximations can be used for the log survival function, which are more accurate than `1 - cdf(x)` when `x >> 1`.
Parameters
ndarray value
`float` or `double` `Tensor`.
string name
Python `str` prepended to names of ops created by this function.
Returns
Tensor
`Tensor` of shape `sample_shape(x) + self.batch_shape` with values of type `self.dtype`.

object prob(int value, string name)

Probability density/mass function.
Parameters
int value
`float` or `double` `Tensor`.
string name
Python `str` prepended to names of ops created by this function.
Returns
object

object prob(IEnumerable<double> value, string name)

Probability density/mass function.
Parameters
IEnumerable<double> value
`float` or `double` `Tensor`.
string name
Python `str` prepended to names of ops created by this function.
Returns
object

object prob(ndarray value, string name)

Probability density/mass function.
Parameters
ndarray value
`float` or `double` `Tensor`.
string name
Python `str` prepended to names of ops created by this function.
Returns
object

object prob(double value, string name)

Probability density/mass function.
Parameters
double value
`float` or `double` `Tensor`.
string name
Python `str` prepended to names of ops created by this function.
Returns
object

object prob(IGraphNodeBase value, string name)

Probability density/mass function.
Parameters
IGraphNodeBase value
`float` or `double` `Tensor`.
string name
Python `str` prepended to names of ops created by this function.
Returns
object

object prob_dyn(object value, ImplicitContainer<T> name)

Probability density/mass function.
Parameters
object value
`float` or `double` `Tensor`.
ImplicitContainer<T> name
Python `str` prepended to names of ops created by this function.
Returns
object

object sample(IGraphNodeBase sample_shape, Nullable<int> seed, string name)

Generate samples of the specified shape.

Note that a call to `sample()` without arguments will generate a single sample.
Parameters
IGraphNodeBase sample_shape
0D or 1D `int32` `Tensor`. Shape of the generated samples.
Nullable<int> seed
Python integer seed for RNG
string name
name to give to the op.
Returns
object

object sample(int sample_shape, Nullable<int> seed, string name)

Generate samples of the specified shape.

Note that a call to `sample()` without arguments will generate a single sample.
Parameters
int sample_shape
0D or 1D `int32` `Tensor`. Shape of the generated samples.
Nullable<int> seed
Python integer seed for RNG
string name
name to give to the op.
Returns
object

object sample(ImplicitContainer<T> sample_shape, Nullable<int> seed, string name)

Generate samples of the specified shape.

Note that a call to `sample()` without arguments will generate a single sample.
Parameters
ImplicitContainer<T> sample_shape
0D or 1D `int32` `Tensor`. Shape of the generated samples.
Nullable<int> seed
Python integer seed for RNG
string name
name to give to the op.
Returns
object

object sample(ValueTuple<int, object> sample_shape, Nullable<int> seed, string name)

Generate samples of the specified shape.

Note that a call to `sample()` without arguments will generate a single sample.
Parameters
ValueTuple<int, object> sample_shape
0D or 1D `int32` `Tensor`. Shape of the generated samples.
Nullable<int> seed
Python integer seed for RNG
string name
name to give to the op.
Returns
object

object sample(IEnumerable<int> sample_shape, Nullable<int> seed, string name)

Generate samples of the specified shape.

Note that a call to `sample()` without arguments will generate a single sample.
Parameters
IEnumerable<int> sample_shape
0D or 1D `int32` `Tensor`. Shape of the generated samples.
Nullable<int> seed
Python integer seed for RNG
string name
name to give to the op.
Returns
object

object sample_dyn(ImplicitContainer<T> sample_shape, object seed, ImplicitContainer<T> name)

Generate samples of the specified shape.

Note that a call to `sample()` without arguments will generate a single sample.
Parameters
ImplicitContainer<T> sample_shape
0D or 1D `int32` `Tensor`. Shape of the generated samples.
object seed
Python integer seed for RNG
ImplicitContainer<T> name
name to give to the op.
Returns
object

Public static methods

Bernoulli NewDyn(object logits, object probs, ImplicitContainer<T> dtype, ImplicitContainer<T> validate_args, ImplicitContainer<T> allow_nan_stats, ImplicitContainer<T> name)

Initialize Categorical distributions using class log-probabilities. (deprecated)

Warning: THIS FUNCTION IS DEPRECATED. It will be removed after 2019-01-01. Instructions for updating: The TensorFlow Distributions library has moved to TensorFlow Probability (https://github.com/tensorflow/probability). You should update all references to use `tfp.distributions` instead of tf.distributions.
Parameters
object logits
An N-D `Tensor`, `N >= 1`, representing the log probabilities of a set of Categorical distributions. The first `N - 1` dimensions index into a batch of independent distributions and the last dimension represents a vector of logits for each class. Only one of `logits` or `probs` should be passed in.
object probs
An N-D `Tensor`, `N >= 1`, representing the probabilities of a set of Categorical distributions. The first `N - 1` dimensions index into a batch of independent distributions and the last dimension represents a vector of probabilities for each class. Only one of `logits` or `probs` should be passed in.
ImplicitContainer<T> dtype
The type of the event samples (default: int32).
ImplicitContainer<T> validate_args
Python `bool`, default `False`. When `True` distribution parameters are checked for validity despite possibly degrading runtime performance. When `False` invalid inputs may silently render incorrect outputs.
ImplicitContainer<T> allow_nan_stats
Python `bool`, default `True`. When `True`, statistics (e.g., mean, mode, variance) use the value "`NaN`" to indicate the result is undefined. When `False`, an exception is raised if one or more of the statistic's batch members are undefined.
ImplicitContainer<T> name
Python `str` name prefixed to Ops created by this class.

Public properties

object allow_nan_stats get;

object allow_nan_stats_dyn get;

TensorShape batch_shape get;

object batch_shape_dyn get;

object dtype get;

object dtype_dyn get;

TensorShape event_shape get;

object event_shape_dyn get;

object logits get;

Log-odds of a `1` outcome (vs `0`).

object logits_dyn get;

Log-odds of a `1` outcome (vs `0`).

string name get;

object name_dyn get;

IDictionary<object, object> parameters get;

object parameters_dyn get;

Tensor probs get;

Probability of a `1` outcome (vs `0`).

object probs_dyn get;

Probability of a `1` outcome (vs `0`).

object PythonObject get;

object reparameterization_type get;

object reparameterization_type_dyn get;

object validate_args get;

object validate_args_dyn get;