diff --git a/doc/pub/How2ReadData/html/How2ReadData-bs.html b/doc/pub/How2ReadData/html/How2ReadData-bs.html
index 2fe4cce75..07d8216f7 100644
--- a/doc/pub/How2ReadData/html/How2ReadData-bs.html
+++ b/doc/pub/How2ReadData/html/How2ReadData-bs.html
@@ -166,7 +166,7 @@ MathJax.Hub.Config({
 <center>[2] <b>Department of Physics and Astronomy and National Superconducting Cyclotron Laboratory, Michigan State University</b></center>
 <br>
 <p>
-<center><h4>Dec 20, 2019</h4></center> <!-- date -->
+<center><h4>Dec 22, 2019</h4></center> <!-- date -->
 <br>
 <p>
 </div> <!-- end jumbotron -->
@@ -1205,7 +1205,7 @@ Finally, another cost function is the Huber cost function used in robust regress
 It is less sensitive to outliers in data than the squared error cost function.
 A variant for classification is also sometimes used, a quantity we will meet later. 
 $$
-L_{\delta }(a)={\begin{cases}{\frac {1}{2}}{a^{2}}&{\text{for }}|a|\leq \delta ,\\\delta (|a|-{\frac {1}{2}}\delta ),&{\text{otherwise.}}\end{cases}}}L_{\delta }(a)={\begin{cases}{\frac  {1}{2}}{a^{2}}&{\text{for }}|a|\leq \delta ,\\\delta (|a|-{\frac  {1}{2}}\delta ),&{\text{otherwise.}}\end{cases}}
+L_{\delta }(a)={\begin{cases}{\frac {1}{2}}{a^{2}}&{\text{for }}|a|\leq \delta ,\\\delta (|a|-{\frac {1}{2}}\delta ),&{\text{otherwise.}}\end{cases}}}.
 $$
 
 <p>
diff --git a/doc/pub/How2ReadData/html/How2ReadData-reveal.html b/doc/pub/How2ReadData/html/How2ReadData-reveal.html
index 57f2f2221..d2dcaca67 100644
--- a/doc/pub/How2ReadData/html/How2ReadData-reveal.html
+++ b/doc/pub/How2ReadData/html/How2ReadData-reveal.html
@@ -148,7 +148,7 @@ MathJax.Hub.Config({
 <center>[2] <b>Department of Physics and Astronomy and National Superconducting Cyclotron Laboratory, Michigan State University</b></center>
 <br>
 <p>&nbsp;<br>
-<center><h4>Dec 20, 2019</h4></center> <!-- date -->
+<center><h4>Dec 22, 2019</h4></center> <!-- date -->
 <br>
 
 <h2 id="___sec0">Introduction </h2>
@@ -1231,7 +1231,7 @@ It is less sensitive to outliers in data than the squared error cost function.
 A variant for classification is also sometimes used, a quantity we will meet later. 
 <p>&nbsp;<br>
 $$
-L_{\delta }(a)={\begin{cases}{\frac {1}{2}}{a^{2}}&{\text{for }}|a|\leq \delta ,\\\delta (|a|-{\frac {1}{2}}\delta ),&{\text{otherwise.}}\end{cases}}}L_{\delta }(a)={\begin{cases}{\frac  {1}{2}}{a^{2}}&{\text{for }}|a|\leq \delta ,\\\delta (|a|-{\frac  {1}{2}}\delta ),&{\text{otherwise.}}\end{cases}}
+L_{\delta }(a)={\begin{cases}{\frac {1}{2}}{a^{2}}&{\text{for }}|a|\leq \delta ,\\\delta (|a|-{\frac {1}{2}}\delta ),&{\text{otherwise.}}\end{cases}}}.
 $$
 <p>&nbsp;<br>
 
diff --git a/doc/pub/How2ReadData/html/How2ReadData-solarized.html b/doc/pub/How2ReadData/html/How2ReadData-solarized.html
index 2d5f087e3..9d93a5ce4 100644
--- a/doc/pub/How2ReadData/html/How2ReadData-solarized.html
+++ b/doc/pub/How2ReadData/html/How2ReadData-solarized.html
@@ -135,7 +135,7 @@ MathJax.Hub.Config({
 <center>[2] <b>Department of Physics and Astronomy and National Superconducting Cyclotron Laboratory, Michigan State University</b></center>
 <br>
 <p>
-<center><h4>Dec 20, 2019</h4></center> <!-- date -->
+<center><h4>Dec 22, 2019</h4></center> <!-- date -->
 <br>
 
 <h2 id="___sec0">Introduction </h2>
@@ -1165,7 +1165,7 @@ Finally, another cost function is the Huber cost function used in robust regress
 It is less sensitive to outliers in data than the squared error cost function.
 A variant for classification is also sometimes used, a quantity we will meet later. 
 $$
-L_{\delta }(a)={\begin{cases}{\frac {1}{2}}{a^{2}}&{\text{for }}|a|\leq \delta ,\\\delta (|a|-{\frac {1}{2}}\delta ),&{\text{otherwise.}}\end{cases}}}L_{\delta }(a)={\begin{cases}{\frac  {1}{2}}{a^{2}}&{\text{for }}|a|\leq \delta ,\\\delta (|a|-{\frac  {1}{2}}\delta ),&{\text{otherwise.}}\end{cases}}
+L_{\delta }(a)={\begin{cases}{\frac {1}{2}}{a^{2}}&{\text{for }}|a|\leq \delta ,\\\delta (|a|-{\frac {1}{2}}\delta ),&{\text{otherwise.}}\end{cases}}}.
 $$
 
 <p>
diff --git a/doc/pub/How2ReadData/html/How2ReadData.html b/doc/pub/How2ReadData/html/How2ReadData.html
index 26137487f..837ec179d 100644
--- a/doc/pub/How2ReadData/html/How2ReadData.html
+++ b/doc/pub/How2ReadData/html/How2ReadData.html
@@ -140,7 +140,7 @@ MathJax.Hub.Config({
 <center>[2] <b>Department of Physics and Astronomy and National Superconducting Cyclotron Laboratory, Michigan State University</b></center>
 <br>
 <p>
-<center><h4>Dec 20, 2019</h4></center> <!-- date -->
+<center><h4>Dec 22, 2019</h4></center> <!-- date -->
 <br>
 
 <h2 id="___sec0">Introduction </h2>
@@ -1170,7 +1170,7 @@ Finally, another cost function is the Huber cost function used in robust regress
 It is less sensitive to outliers in data than the squared error cost function.
 A variant for classification is also sometimes used, a quantity we will meet later. 
 $$
-L_{\delta }(a)={\begin{cases}{\frac {1}{2}}{a^{2}}&{\text{for }}|a|\leq \delta ,\\\delta (|a|-{\frac {1}{2}}\delta ),&{\text{otherwise.}}\end{cases}}}L_{\delta }(a)={\begin{cases}{\frac  {1}{2}}{a^{2}}&{\text{for }}|a|\leq \delta ,\\\delta (|a|-{\frac  {1}{2}}\delta ),&{\text{otherwise.}}\end{cases}}
+L_{\delta }(a)={\begin{cases}{\frac {1}{2}}{a^{2}}&{\text{for }}|a|\leq \delta ,\\\delta (|a|-{\frac {1}{2}}\delta ),&{\text{otherwise.}}\end{cases}}}.
 $$
 
 <p>
diff --git a/doc/pub/How2ReadData/ipynb/How2ReadData.ipynb b/doc/pub/How2ReadData/ipynb/How2ReadData.ipynb
index dd60965d9..51081a4c6 100644
--- a/doc/pub/How2ReadData/ipynb/How2ReadData.ipynb
+++ b/doc/pub/How2ReadData/ipynb/How2ReadData.ipynb
@@ -10,7 +10,7 @@
     "<!-- Author: -->  \n",
     "**Morten Hjorth-Jensen**, Department of Physics, University of Oslo and Department of Physics and Astronomy and National Superconducting Cyclotron Laboratory, Michigan State University\n",
     "\n",
-    "Date: **Dec 20, 2019**\n",
+    "Date: **Dec 22, 2019**\n",
     "\n",
     "Copyright 1999-2019, Morten Hjorth-Jensen. Released under CC Attribution-NonCommercial 4.0 license\n",
     "\n",
@@ -1467,7 +1467,7 @@
    "metadata": {},
    "source": [
     "$$\n",
-    "L_{\\delta }(a)={\\begin{cases}{\\frac {1}{2}}{a^{2}}&{\\text{for }}|a|\\leq \\delta ,\\\\\\delta (|a|-{\\frac {1}{2}}\\delta ),&{\\text{otherwise.}}\\end{cases}}}L_{\\delta }(a)={\\begin{cases}{\\frac  {1}{2}}{a^{2}}&{\\text{for }}|a|\\leq \\delta ,\\\\\\delta (|a|-{\\frac  {1}{2}}\\delta ),&{\\text{otherwise.}}\\end{cases}}\n",
+    "L_{\\delta }(a)={\\begin{cases}{\\frac {1}{2}}{a^{2}}&{\\text{for }}|a|\\leq \\delta ,\\\\\\delta (|a|-{\\frac {1}{2}}\\delta ),&{\\text{otherwise.}}\\end{cases}}}.\n",
     "$$"
    ]
   },
diff --git a/doc/pub/How2ReadData/ipynb/ipynb-How2ReadData-src.tar.gz b/doc/pub/How2ReadData/ipynb/ipynb-How2ReadData-src.tar.gz
index c22a7cf62..cf8fc4d65 100644
Binary files a/doc/pub/How2ReadData/ipynb/ipynb-How2ReadData-src.tar.gz and b/doc/pub/How2ReadData/ipynb/ipynb-How2ReadData-src.tar.gz differ
diff --git a/doc/pub/How2ReadData/pdf/How2ReadData-minted.pdf b/doc/pub/How2ReadData/pdf/How2ReadData-minted.pdf
index 92658c57d..6620ab57c 100644
Binary files a/doc/pub/How2ReadData/pdf/How2ReadData-minted.pdf and b/doc/pub/How2ReadData/pdf/How2ReadData-minted.pdf differ
diff --git a/doc/src/How2ReadData/How2ReadData.do.txt b/doc/src/How2ReadData/How2ReadData.do.txt
index 68c2b07c4..c797c99a3 100644
--- a/doc/src/How2ReadData/How2ReadData.do.txt
+++ b/doc/src/How2ReadData/How2ReadData.do.txt
@@ -942,7 +942,7 @@ It is less sensitive to outliers in data than the squared error cost function.
 A variant for classification is also sometimes used, a quantity we will meet later. 
 !bt
 \[
-L_{\delta }(a)={\begin{cases}{\frac {1}{2}}{a^{2}}&{\text{for }}|a|\leq \delta ,\\\delta (|a|-{\frac {1}{2}}\delta ),&{\text{otherwise.}}\end{cases}}}L_{\delta }(a)={\begin{cases}{\frac  {1}{2}}{a^{2}}&{\text{for }}|a|\leq \delta ,\\\delta (|a|-{\frac  {1}{2}}\delta ),&{\text{otherwise.}}\end{cases}}
+L_{\delta }(a)={\begin{cases}{\frac {1}{2}}{a^{2}}&{\text{for }}|a|\leq \delta ,\\\delta (|a|-{\frac {1}{2}}\delta ),&{\text{otherwise.}}\end{cases}}}.
 \]
 !et