Merge pull request #3050 from madeline-underwood/eigen

Update Eigen documentation for consistency and clarity in headings

Author: jasonrandrews

content/learning-paths/cross-platform/eigen-linear-algebra-on-arm/_index.md

@@ -1,5 +1,8 @@
 ---
 title: Use the Eigen Linear Algebra Library on Arm
+
+description: Learn how to use the Eigen linear algebra library on Arm systems with ASIMD and SVE vectorization, including building TensorFlow with SVE support for optimized performance.
+
 minutes_to_complete: 45
 
 who_is_this_for: This is an advanced topic for C/C++ developers who want to create high performance applications using the Eigen linear algebra library.

content/learning-paths/cross-platform/eigen-linear-algebra-on-arm/eigen-intro.md

@@ -104,7 +104,7 @@ There are more specialized classes for both storage types, like the `Tensor` cla
 
 In this Learning Path, you will test Eigen using a few examples with emphasis on Arm CPUs. For full documentation and a list of all supported types and operations on them, see the [Eigen documentation](https://libeigen.gitlab.io/docs/).
 
-## Numerical Solvers
+## Numerical solvers
 
 Besides basic vector and matrix types, Eigen also provides built-in methods for numerical solving on the matrices.
 

content/learning-paths/cross-platform/eigen-linear-algebra-on-arm/eigen-on-arm-part1.md

@@ -8,7 +8,7 @@ layout: learningpathall
 
 You have learned that Eigen has been ported to Neon/ASIMD in the past and recently to SVE. Now you are going to investigate how to use Eigen to get best performance on Arm systems with SIMD engines.
 
-## Example 1: Arbitrary-Size Matrix Sum of all Elements
+## Example 1: arbitrary-size matrix sum of all elements
 
 This example demonstrates the benefits of Eigen's vectorization with some simple expressions on matrices. It first constructs a large random matrix of 100 x 100 `float` elements, and performs some simple operation for `N` iterations on it. Finally, it returns the sum of all the elements of the final matrix as a `double`.
 

content/learning-paths/cross-platform/eigen-linear-algebra-on-arm/eigen-on-arm-part2.md

@@ -6,7 +6,7 @@ weight: 4
 layout: learningpathall
 ---
 
-## Example 2: Matrix Multiplication
+## Example 2: matrix multiplication
 
 Use a text editor to save the program below in a file named `eigen-test3.cpp`:
 
@@ -81,7 +81,7 @@ You again see that the SVE build is slower than the ASIMD.
 
 The SVE implementation is still experimental and performance is expected to improve in the future.
 
-## Example 3: Vector 3D Rotation
+## Example 3: vector 3D rotation
 
 This example creates an initial random normalized vector `v1`.