public:
  explicit SqrtGradientFunction(AbstractTensorHandle* sqrt) : sqrt_(sqrt) {
    sqrt->Ref();
  }
  Status Compute(AbstractContext* ctx,
                 absl::Span<AbstractTensorHandle* const> grad_outputs,
                 absl::Span<AbstractTensorHandle*> grad_inputs) override {
    std::string name = "Sqrt_Grad";
    TF_RETURN_IF_ERROR(SqrtGrad(ctx, sqrt_.get(), grad_outputs[0],

}

Status SqrtModel(AbstractContext* ctx,
                 absl::Span<AbstractTensorHandle* const> inputs,
                 absl::Span<AbstractTensorHandle*> outputs) {
  return ops::Sqrt(ctx, inputs[0], &outputs[0], "Sqrt");
}

Status NegModel(AbstractContext* ctx,
                absl::Span<AbstractTensorHandle* const> inputs,
                absl::Span<AbstractTensorHandle*> outputs) {

        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))`