For example, if two backends are specified with equal weights, and one is
                        invalid, 50 percent of traffic MUST receive an `UNAVAILABLE` status.
                        Implementations may choose how that 50 percent is determined.


                        Support: Core for Kubernetes Service

	FACILITY_FVE                                                                            = 49
	FACILITY_FWP                                                                            = 50
	FACILITY_WINRM                                                                          = 51
	FACILITY_NDIS                                                                           = 52

value is computed as \\( \sqrt{a^2 + b^2}\\).

For example:

>>> x = tf.complex(3.0, 4.0)
>>> print((tf.raw_ops.ComplexAbs(x=x, Tout=tf.dtypes.float32, name=None)).numpy())
5.0
  }];

  let arguments = (ins
    TensorOf<[TF_Complex128, TF_Complex64]>:$x
  );

  let results = (outs
    TF_F32OrF64Tensor:$y
  );

		v0.AddArg2(x, x)
		v.AddArg(v0)
		return true
	}
	// match: (MULLconst [c] x)
	// cond: c%5 == 0 && isPowerOfTwo32(c/5)
	// result: (SHLLconst [int8(log32(c/5))] (LEAL4 <v.Type> x x))
	for {
		c := auxIntToInt32(v.AuxInt)
		x := v_0
		if !(c%5 == 0 && isPowerOfTwo32(c/5)) {
			break
		}
		v.reset(OpAMD64SHLLconst)
		v.AuxInt = int8ToAuxInt(int8(log32(c / 5)))

    modules are under `tf.lite` and source code is now under `tensorflow/lite`
    rather than `tensorflow/contrib/lite`.
*   TensorFlow GPU binaries are now built against CUDA 10 and TensorRT 5.0.
*   Support for Python3.7 on all operating systems.
*   Moved NCCL to core.

## Behavioral changes

*   Disallow conversion of python floating types to uint32/64 (matching behavior