added about huber cost function

doc/pub/How2ReadData/ipynb/How2ReadData.ipynb

@@ -1459,10 +1459,22 @@
     "\n",
     "Finally, another cost function is the Huber cost function used in robust regression.\n",
     "It is less sensitive to outliers in data than the squared error cost function.\n",
-    "A variant for classification is also sometimes used, a quantity we will meet later. \n",
-    "\n",
-    "\n",
-    "\n",
+    "A variant for classification is also sometimes used, a quantity we will meet later."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "$$\n",
+    "L_{\\delta }(a)={\\begin{cases}{\\frac {1}{2}}{a^{2}}&{\\text{for }}|a|\\leq \\delta ,\\\\\\delta (|a|-{\\frac {1}{2}}\\delta ),&{\\text{otherwise.}}\\end{cases}}}L_{\\delta }(a)={\\begin{cases}{\\frac  {1}{2}}{a^{2}}&{\\text{for }}|a|\\leq \\delta ,\\\\\\delta (|a|-{\\frac  {1}{2}}\\delta ),&{\\text{otherwise.}}\\end{cases}}\n",
+    "$$"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
     "We will discuss in more\n",
     "detail these and other functions in the various lectures.  We conclude this part with another example. Instead of \n",
     "a linear $x$-dependence we study now a cubic polynomial and use the polynomial regression analysis tools of scikit-learn."