[ci skip] fix documentation

documentation/release_6.3.htm

@@ -50,7 +50,7 @@ <h4>General</h4>
       The priors code has been refactored to provide two common, parallelized base classes for CPU and GPU implementations (<code>GibbsPenalty</code> and <code>CudaGibbsPenalty</code>).<br>
       On CPU, we introduce <code>GibbsQuadraticPenalty</code> and <code>GibbsRelativeDifferencePenalty</code>, which inherit from the parallelized <code>GibbsPenalty</code> base class (now parallelized using OpenMP).<br>
       GPU implementations are also provided as <code>CudaGibbsQuadraticPenalty</code> and <code>CudaGibbsRelativeDifferencePenalty</code>.<br>
-      Currently, the new Gibbs penalties are still missing the paraboloidal surrogates related method implementations. They provide two new methods: <code>compute_gradient_times_input</code> and <code>compute_Hessian_times_input</code>.<br>
+      Currently, the new Gibbs penalties are still missing the paraboloidal surrogates related method implementations. They provide two new methods: <code>compute_gradient_times_input</code> and <code>compute_Hessian_diagonalt</code>.<br>
       <a href="https://github.com/UCL/STIR/pull/1629">PR #1629</a>
     </li>
     <li>
@@ -258,7 +258,7 @@ <h3>New Deprecations and renames</h3>
       <tt>STIR_ROOT_ROTATION_AS_V4</tt> will therefore be removed.
     </li>
     <li>
-      Starting from v7.0, <code>GeneralizedPriors</code> will be renamed to <code>GeneralizedPenalty</code>.<br>
+      Starting from v7.0, <code>GeneralizedPrior</code> will be renamed to <code>GeneralizedPenalty</code>.<br>
       <code>QuadraticPrior</code>, <code>RelativeDifferencePrior</code>, and <code>LogcoshPrior</code> will be removed in favor of the new <code>GibbsPenalty</code> framework.
       <a href=https://github.com/UCL/STIR/issues/1426>Issue #1426</a>
     </li>