[SPARK-30995][ML][DOCS] Latex doesn't work correctly in FMClassifier/…

…FMRegressor Scala doc ### What changes were proposed in this pull request? Latex doesn't work correctly ### Why are the changes needed? Fix the doc to make Latex work ### Does this PR introduce any user-facing change? Before fix: ![image](https://user-images.githubusercontent.com/13592258/75611743-0fa00a00-5ad2-11ea-9cc0-4b246b25d63d.png) ![image](https://user-images.githubusercontent.com/13592258/75611755-25adca80-5ad2-11ea-884d-b2792b714bd5.png) After fix: ![image](https://user-images.githubusercontent.com/13592258/75611776-46762000-5ad2-11ea-838d-f7f6f93c8aec.png) ![image](https://user-images.githubusercontent.com/13592258/75611778-51c94b80-5ad2-11ea-85e4-8c7424268f52.png) ### How was this patch tested? Manually build doc and test Closes apache#27748 from huaxingao/fm_doc. Authored-by: Huaxin Gao <[email protected]> Signed-off-by: HyukjinKwon <[email protected]>
andremoeller · Mar 2, 2020 · 0a5e9a1 · 0a5e9a1
1 parent f828453
commit 0a5e9a1
Show file tree

Hide file tree

Showing 2 changed files with 14 additions and 6 deletions.
diff --git a/mllib/src/main/scala/org/apache/spark/ml/classification/FMClassifier.scala b/mllib/src/main/scala/org/apache/spark/ml/classification/FMClassifier.scala
@@ -51,12 +51,16 @@ private[classification] trait FMClassifierParams extends ProbabilisticClassifier
  * FM is able to estimate interactions even in problems with huge sparsity
  * (like advertising and recommendation system).
  * FM formula is:
- * {{{
+ * <blockquote>
+ *   $$
+ *   \begin{align}
  *   y = \sigma\left( w_0 + \sum\limits^n_{i-1} w_i x_i +
  *     \sum\limits^n_{i=1} \sum\limits^n_{j=i+1} \langle v_i, v_j \rangle x_i x_j \right)
- * }}}
+ *   \end{align}
+ *   $$
+ * </blockquote>
  * First two terms denote global bias and linear term (as same as linear regression),
- * and last term denotes pairwise interactions term. {{{v_i}}} describes the i-th variable
+ * and last term denotes pairwise interactions term. v_i describes the i-th variable
  * with k factors.
  *
  * FM classification model uses logistic loss which can be solved by gradient descent method, and

diff --git a/mllib/src/main/scala/org/apache/spark/ml/regression/FMRegressor.scala b/mllib/src/main/scala/org/apache/spark/ml/regression/FMRegressor.scala
@@ -276,12 +276,16 @@ private[regression] trait FMRegressorParams extends FactorizationMachinesParams
  * FM is able to estimate interactions even in problems with huge sparsity
  * (like advertising and recommendation system).
  * FM formula is:
- * {{{
+ * <blockquote>
+ *   $$
+ *   \begin{align}
  *   y = w_0 + \sum\limits^n_{i-1} w_i x_i +
  *     \sum\limits^n_{i=1} \sum\limits^n_{j=i+1} \langle v_i, v_j \rangle x_i x_j
- * }}}
+ *   \end{align}
+ *   $$
+ * </blockquote>
  * First two terms denote global bias and linear term (as same as linear regression),
- * and last term denotes pairwise interactions term. {{{v_i}}} describes the i-th variable
+ * and last term denotes pairwise interactions term. v_i describes the i-th variable
  * with k factors.
  *
  * FM regression model uses MSE loss which can be solved by gradient descent method, and