*   Enable JIT-compiled i64-indexed kernels on GPU for large tensors with more than 2**32 elements.
    *   Unary GPU kernels: Abs, Atanh, Acos, Acosh, Asin, Asinh, Atan, Cos, Cosh, Sin, Sinh, Tan, Tanh.
    *   Binary GPU kernels: AddV2, Sub, Div, DivNoNan, Mul, MulNoNan, FloorDiv, Equal, NotEqual, Greater, GreaterEqual, LessEqual, Less.

* `tf.lite`

The `tf.math.atan` operation returns the inverse of `tf.math.tan`, such that
if `y = tf.math.tan(x)` then, `x = tf.math.atan(y)`.

**Note**: The output of `tf.math.atan` will lie within the invertible range
of tan, i.e (-pi/2, pi/2).

For example:

```python
# Note: [1.047, 0.785] ~= [(pi/3), (pi/4)]
x = tf.constant([1.047, 0.785])
y = tf.math.tan(x) # [1.731261, 0.99920404]

tf.math.atan(y) # [1.047, 0.785] = x

		{"SmallestNonzeroFloat64", Const, 0},
		{"Sqrt", Func, 0},
		{"Sqrt2", Const, 0},
		{"SqrtE", Const, 0},
		{"SqrtPhi", Const, 0},
		{"SqrtPi", Const, 0},
		{"Tan", Func, 0},
		{"Tanh", Func, 0},
		{"Trunc", Func, 0},
		{"Y0", Func, 0},
		{"Y1", Func, 0},
		{"Yn", Func, 0},
	},
	"math/big": {
		{"(*Float).Abs", Method, 5},
		{"(*Float).Acc", Method, 5},