}];

  let description = [{
The input is a tensor of shape `[..., M, M]` whose inner-most 2 dimensions
form square matrices.

The input has to be symmetric and positive definite. Only the lower-triangular
part of the input will be used for this operation. The upper-triangular part
will not be read.

The output is a tensor of the same shape as the input

2066..2069    ; disallowed                             # 6.3  LEFT-TO-RIGHT ISOLATE..POP DIRECTIONAL ISOLATE
206A..206F    ; disallowed                             # 1.1  INHIBIT SYMMETRIC SWAPPING..NOMINAL DIGIT SHAPES
2070          ; mapped                 ; 0030          # 1.1  SUPERSCRIPT ZERO
2071          ; mapped                 ; 0069          # 3.2  SUPERSCRIPT LATIN SMALL LETTER I

        `TFLiteConverter.from_saved_model`.
    *   Added DEPTH_TO_SPACE support in Post training quantization.
    *   Added dynamic range quantization support for the BatchMatMul op.
        *   Both symmetric and asymmetric quantized input tensor are supported.
    *   Add `RFFT2D` as builtin op. (`RFFT2D` also supports `RFFTD`.) Currently
        only supports float32 input.
    *   Add 5D support to `SLICE` op.