Function case

tensorflow/python/ops/control_flow_ops.py:3386–3490

Create a case operation. See also `tf.switch_case`. The `pred_fn_pairs` parameter is a dict or list of pairs of size N. Each pair contains a boolean scalar tensor and a python callable that creates the tensors to be returned if the boolean evaluates to True. `default` is a callable gener

(pred_fn_pairs,
         default=None,
         exclusive=False,
         strict=False,
         name="case")

@tf_export(v1=["case"])
def case(pred_fn_pairs,
         default=None,
         exclusive=False,
         strict=False,
         name="case"):
  """Create a case operation.

  See also `tf.switch_case`.

  The `pred_fn_pairs` parameter is a dict or list of pairs of size N.
  Each pair contains a boolean scalar tensor and a python callable that
  creates the tensors to be returned if the boolean evaluates to True.
  `default` is a callable generating a list of tensors. All the callables
  in `pred_fn_pairs` as well as `default` (if provided) should return the same
  number and types of tensors.

  If `exclusive==True`, all predicates are evaluated, and an exception is
  thrown if more than one of the predicates evaluates to `True`.
  If `exclusive==False`, execution stops at the first predicate which
  evaluates to True, and the tensors generated by the corresponding function
  are returned immediately. If none of the predicates evaluate to True, this
  operation returns the tensors generated by `default`.

  `tf.case` supports nested structures as implemented in
  `tf.contrib.framework.nest`. All of the callables must return the same
  (possibly nested) value structure of lists, tuples, and/or named tuples.
  Singleton lists and tuples form the only exceptions to this: when returned by
  a callable, they are implicitly unpacked to single values. This
  behavior is disabled by passing `strict=True`.

  If an unordered dictionary is used for `pred_fn_pairs`, the order of the
  conditional tests is not guaranteed. However, the order is guaranteed to be
  deterministic, so that variables created in conditional branches are created
  in fixed order across runs.

  @compatibility(eager)
  Unordered dictionaries are not supported in eager mode when `exclusive=False`.
  Use a list of tuples instead.
  @end_compatibility


  **Example 1:**

  Pseudocode:

  ```
  if (x < y) return 17;
  else return 23;
  ```

  Expressions:

  ```python
  f1 = lambda: tf.constant(17)
  f2 = lambda: tf.constant(23)
  r = tf.case([(tf.less(x, y), f1)], default=f2)
  ```