Fix some typos (distribuion, funciotn, etc.) (PaddlePaddle#62000)

python/paddle/distribution/binomial.py

@@ -116,7 +116,7 @@ def _check_constraint(self, total_count, probs):
 
     @property
     def mean(self):
-        """Mean of binomial distribuion.
+        """Mean of binomial distribution.
 
         Returns:
             Tensor: mean value.

python/paddle/distribution/continuous_bernoulli.py

@@ -210,7 +210,7 @@ def _log_constant(self):
 
     @property
     def mean(self):
-        """Mean of Continuous Bernoulli distribuion.
+        """Mean of Continuous Bernoulli distribution.
 
         Returns:
             Tensor: mean value.

python/paddle/distribution/dirichlet.py

@@ -51,7 +51,7 @@ class Dirichlet(exponential_family.ExponentialFamily):
             distribution, also called :math:`\alpha`. When it's over one
             dimension, the last axis denotes the parameter of distribution,
             ``event_shape=concentration.shape[-1:]`` , axes other than last are
-            condsider batch dimensions with ``batch_shape=concentration.shape[:-1]`` .
+            consider batch dimensions with ``batch_shape=concentration.shape[:-1]`` .
 
     Examples:
 
@@ -79,7 +79,7 @@ def __init__(self, concentration):
 
     @property
     def mean(self):
-        """Mean of Dirichelt distribution.
+        """Mean of Dirichlet distribution.
 
         Returns:
             Mean value of distribution.

python/paddle/distribution/exponential.py

@@ -83,7 +83,7 @@ def __init__(self, rate):
 
     @property
     def mean(self):
-        """Mean of exponential distribuion.
+        """Mean of exponential distribution.
 
         Returns:
             Tensor: mean value.
@@ -134,7 +134,7 @@ def rsample(self, shape=()):
         return -paddle.log(uniform) / self.rate
 
     def prob(self, value):
-        r"""Probability density funciotn evaluated at value.
+        r"""Probability density function evaluated at value.
 
         .. math::
 

python/paddle/distribution/exponential_family.py

@@ -48,7 +48,7 @@ def _mean_carrier_measure(self):
         raise NotImplementedError
 
     def entropy(self):
-        """caculate entropy use `bregman divergence`
+        """calculate entropy use `bregman divergence`
         https://www.lix.polytechnique.fr/~nielsen/EntropyEF-ICIP2010.pdf
         """
         entropy_value = -self._mean_carrier_measure

python/paddle/distribution/gamma.py

@@ -113,7 +113,7 @@ def __init__(self, concentration, rate):
 
     @property
     def mean(self):
-        """Mean of gamma distribuion.
+        """Mean of gamma distribution.
 
         Returns:
             Tensor: mean value.
@@ -130,7 +130,7 @@ def variance(self):
         return self.concentration / self.rate.pow(2)
 
     def prob(self, value):
-        """Probability density funciotn evaluated at value
+        """Probability density function evaluated at value
 
         Args:
             value (float|Tensor): Value to be evaluated.

python/paddle/distribution/geometric.py

@@ -124,7 +124,7 @@ def stddev(self):
         return paddle.sqrt(self.variance)
 
     def pmf(self, k):
-        r"""Probability mass funciotn evaluated at k.
+        r"""Probability mass function evaluated at k.
 
         .. math::
 
@@ -341,5 +341,5 @@ def kl_divergence(self, other):
             )
         else:
             raise TypeError(
-                f"Exected type of other is geometric.Geometric, but got {type(other)}"
+                f"Exacted type of other is geometric.Geometric, but got {type(other)}"
             )

python/paddle/distribution/kl.py

@@ -73,13 +73,13 @@ def kl_divergence(p, q):
 
 
 def register_kl(cls_p, cls_q):
-    """Decorator for register a KL divergence implemention function.
+    """Decorator for register a KL divergence implementation function.
 
-    The ``kl_divergence(p, q)`` function will search concrete implemention
+    The ``kl_divergence(p, q)`` function will search concrete implementation
     functions registered by ``register_kl``, according to multi-dispatch pattern.
-    If an implemention function is found, it will return the result, otherwise,
+    If an implementation function is found, it will return the result, otherwise,
     it will raise ``NotImplementError`` exception. Users can register
-    implemention function by the decorator.
+    implementation function by the decorator.
 
     Args:
         cls_p (Distribution): The Distribution type of Instance p. Subclass derived from ``Distribution``.
@@ -110,16 +110,16 @@ def _dispatch(cls_p, cls_q):
     """Multiple dispatch into concrete implement function."""
 
     # find all matched super class pair of p and q
-    matchs = [
+    matches = [
         (super_p, super_q)
         for super_p, super_q in _REGISTER_TABLE
         if issubclass(cls_p, super_p) and issubclass(cls_q, super_q)
     ]
-    if not matchs:
+    if not matches:
         raise NotImplementedError
 
-    left_p, left_q = min(_Compare(*m) for m in matchs).classes
-    right_p, right_q = min(_Compare(*reversed(m)) for m in matchs).classes
+    left_p, left_q = min(_Compare(*m) for m in matches).classes
+    right_p, right_q = min(_Compare(*reversed(m)) for m in matches).classes
 
     if _REGISTER_TABLE[left_p, left_q] is not _REGISTER_TABLE[right_p, right_q]:
         warnings.warn(

python/paddle/distribution/lognormal.py

@@ -97,7 +97,7 @@ def __init__(self, loc, scale):
 
     @property
     def mean(self):
-        """Mean of lognormal distribuion.
+        """Mean of lognormal distribution.
 
         Returns:
             Tensor: mean value.

python/paddle/distribution/multinomial.py

@@ -87,7 +87,7 @@ def __init__(self, total_count, probs):
 
     @property
     def mean(self):
-        """mean of multinomial distribuion.
+        """mean of multinomial distribution.
 
         Returns:
             Tensor: mean value.

python/paddle/distribution/multivariate_normal.py

@@ -150,7 +150,7 @@ def __init__(
                 batch_shape
                 + [precision_matrix.shape[-2], precision_matrix.shape[-1]]
             )
-        self._check_constriants()
+        self._check_constraints()
         self.loc = loc.expand(
             batch_shape
             + [
@@ -223,16 +223,16 @@ def _check_positive_definite(self, value):
             raise ValueError(
                 "covariance_matrix or precision_matrix must be a symmetric matrix"
             )
-        is_postive_definite = (
+        is_positive_definite = (
             paddle.cast(paddle.linalg.eigvalsh(value), dtype=self.dtype) > 0
         ).all()
-        return is_postive_definite
+        return is_positive_definite
 
-    def _check_constriants(self):
-        """Check whether the matrix satisfy corresponding constriant
+    def _check_constraints(self):
+        """Check whether the matrix satisfy corresponding constraint
 
         Return:
-            Tensor: indicator for the pass of constriants check
+            Tensor: indicator for the pass of constraints check
         """
         if self.scale_tril is not None:
             check = self._check_lower_triangular(self.scale_tril)
@@ -241,25 +241,25 @@ def _check_constriants(self):
                     "scale_tril matrix must be a lower triangular matrix with positive diagonals"
                 )
         elif self.covariance_matrix is not None:
-            is_postive_definite = self._check_positive_definite(
+            is_positive_definite = self._check_positive_definite(
                 self.covariance_matrix
             )
-            if not is_postive_definite:
+            if not is_positive_definite:
                 raise ValueError(
                     "covariance_matrix must be a symmetric positive definite matrix"
                 )
         else:
-            is_postive_definite = self._check_positive_definite(
+            is_positive_definite = self._check_positive_definite(
                 self.precision_matrix
             )
-            if not is_postive_definite:
+            if not is_positive_definite:
                 raise ValueError(
                     "precision_matrix must be a symmetric positive definite matrix"
                 )
 
     @property
     def mean(self):
-        """Mean of Multivariate Normal distribuion.
+        """Mean of Multivariate Normal distribution.
 
         Returns:
             Tensor: mean value.
@@ -451,7 +451,7 @@ def precision_to_scale_tril(P):
 
 def batch_mahalanobis(bL, bx):
     r"""
-    Computes the squared Mahalanobis distance of the Multivariate Normal distribution with cholesky decomposition of the covatiance matrix.
+    Computes the squared Mahalanobis distance of the Multivariate Normal distribution with cholesky decomposition of the covariance matrix.
     Accepts batches for both bL and bx.
 
     Args:

python/paddle/distribution/normal.py

@@ -159,7 +159,7 @@ def __init__(self, loc, scale, name=None):
 
     @property
     def mean(self):
-        """Mean of normal distribuion.
+        """Mean of normal distribution.
 
         Returns:
             Tensor: mean value.

python/paddle/distribution/poisson.py

@@ -112,7 +112,7 @@ def _check_constraint(self, value):
 
     @property
     def mean(self):
-        """Mean of poisson distribuion.
+        """Mean of poisson distribution.
 
         Returns:
             Tensor: mean value.

python/paddle/distribution/transform.py

@@ -131,7 +131,7 @@ def _is_injective(cls):
 
     def __call__(self, input):
         """Make this instance as a callable object. The return value is
-        depening on the input type.
+        depending on the input type.
 
         * If the input is a ``Tensor`` instance, return
           ``self.forward(input)`` .
@@ -160,7 +160,7 @@ def forward(self, x):
         Useful for turning one random outcome into another.
 
         Args:
-            x (Tensos): Input parameter, generally is a sample generated
+            x (Tensor): Input parameter, generally is a sample generated
                 from ``Distribution``.
 
         Returns:
@@ -299,7 +299,7 @@ def _codomain(self):
         return variable.real
 
     def _forward(self, x):
-        """Inner method for publid API ``forward``, subclass should
+        """Inner method for public API ``forward``, subclass should
         overwrite this method for supporting forward transformation.
         """
         raise NotImplementedError('Forward not implemented')
@@ -340,8 +340,8 @@ def _forward_shape(self, shape):
         return shape
 
     def _inverse_shape(self, shape):
-        """Inner method called by ``inverse_shape``, whic is used to infer the
-        invese shape. Subclass should overwrite this method for supporting
+        """Inner method called by ``inverse_shape``, which is used to infer the
+        inverse shape. Subclass should overwrite this method for supporting
         ``inverse_shape``.
         """
         return shape
@@ -355,7 +355,7 @@ class AbsTransform(Transform):
     distributions with the absolute value function, which maps ``(-inf, inf)``
     to ``[0, inf)`` .
 
-    * For ``y`` in ``(0, inf)`` , ``AbsTransform.inverse(y)`` returns the set invese
+    * For ``y`` in ``(0, inf)`` , ``AbsTransform.inverse(y)`` returns the set inverse
       ``{x  in (-inf, inf) : |x| = y}`` as a tuple, ``-y, y`` .
     * For ``y`` equal ``0`` , ``AbsTransform.inverse(0)`` returns ``0, 0``, which is not
       the set inverse (the set inverse is the singleton {0}), but "works" in
@@ -681,13 +681,13 @@ class IndependentTransform(Transform):
     some of the rightmost batch axes as event axes.
 
     Generally, it is used to expand the event axes. This has no effect on the
-    forward or inverse transformaion, but does sum out the
-    ``reinterpretd_bach_rank`` rightmost dimensions in computing the determinant
+    forward or inverse transformation, but does sum out the
+    ``reinterpreted_batch_rank`` rightmost dimensions in computing the determinant
     of Jacobian matrix.
 
     To see this, consider the ``ExpTransform`` applied to a Tensor which has
     sample, batch, and event ``(S,B,E)`` shape semantics. Suppose the Tensor's
-    paritioned-shape is ``(S=[4], B=[2, 2], E=[3])`` , reinterpreted_batch_rank
+    partitioned-shape is ``(S=[4], B=[2, 2], E=[3])`` , reinterpreted_batch_rank
     is 1. Then the reinterpreted Tensor's shape  is ``(S=[4], B=[2], E=[2, 3])`` .
     The shape returned by ``forward`` and ``inverse`` is unchanged, ie,
     ``[4,2,2,3]`` . However the shape returned by ``inverse_log_det_jacobian``
@@ -875,7 +875,7 @@ def __init__(self, in_event_shape, out_event_shape):
         ):
             raise TypeError(
                 f"Expected type of 'in_event_shape' and 'out_event_shape' is "
-                f"Squence[int], but got 'in_event_shape': {in_event_shape}, "
+                f"Sequence[int], but got 'in_event_shape': {in_event_shape}, "
                 f"'out_event_shape': {out_event_shape}"
             )
         in_size = 1