fractionalBuilder.add(x);
        }
      }
    }
    INTEGRAL_DOUBLE_CANDIDATES = integralBuilder.build();
    fractionalBuilder.add(1.414).add(1.415).add(Math.sqrt(2));
    fractionalBuilder.add(5.656).add(5.657).add(4 * Math.sqrt(2));
    for (double d : INTEGRAL_DOUBLE_CANDIDATES) {
      double x = 1 / d;
      if (x != Math.rint(x)) {
        fractionalBuilder.add(x);
      }
    }

        assertEquals(BigIntegerMath.sqrt(valueOf(x), mode), valueOf(IntMath.sqrt(x, mode)));
      }
    }
  }

  /* Relies on the correctness of sqrt(int, FLOOR). */
  @GwtIncompatible // sqrt
  public void testSqrtExactMatchesFloorOrThrows() {
    for (int x : POSITIVE_INTEGER_CANDIDATES) {
      int floor = IntMath.sqrt(x, FLOOR);

        assertEquals(BigIntegerMath.sqrt(valueOf(x), mode), valueOf(LongMath.sqrt(x, mode)));
      }
    }
  }

  /* Relies on the correctness of sqrt(long, FLOOR). */
  @GwtIncompatible // TODO
  public void testSqrtExactMatchesFloorOrThrows() {
    for (long x : POSITIVE_LONG_CANDIDATES) {
      long sqrtFloor = LongMath.sqrt(x, FLOOR);

     * Math.sqrt(k * k) <= Math.sqrt(x) <= Math.sqrt((k + 1) * (k + 1))
     *          since Math.sqrt is monotonic.
     * (long) Math.sqrt(k * k) <= (long) Math.sqrt(x) <= (long) Math.sqrt((k + 1) * (k + 1))
     *          since casting to long is monotonic
     * k <= (long) Math.sqrt(x) <= k + 1
     *          since (long) Math.sqrt(k * k) == k, as checked exhaustively in

	XVMODW		X3, X2, X1	// 410ce374
	XVMODV		X3, X2, X1	// 418ce374
	XVMODBU		X3, X2, X1	// 410ce674
	XVMODHU		X3, X2, X1	// 418ce674
	XVMODWU		X3, X2, X1	// 410ce774
	XVMODVU		X3, X2, X1	// 418ce774

	// [X]VF{SQRT/RECIP/RSQRT}{F/D} instructions
	VFSQRTF		V1, V2		// 22e49c72
	VFSQRTD		V1, V2		// 22e89c72
	VFRECIPF	V1, V2		// 22f49c72
	VFRECIPD	V1, V2		// 22f89c72
	VFRSQRTF	V1, V2		// 22049d72
	VFRSQRTD	V1, V2		// 22089d72

        1e-10))` Alternatively, you can override `convolution_op`: `python class
        StandardizedConv2D(tf.keras.Layer): def convolution_op(self, inputs,
        kernel): mean, var = tf.nn.moments(kernel, axes=[0, 1, 2],
        keepdims=True) # Author code uses std + 1e-5 return
        super().convolution_op(inputs, (kernel - mean) / tf.sqrt(var + 1e-10))`