From 40d2cbf17753282f5e210b1793e4ddf024d9e438 Mon Sep 17 00:00:00 2001
From: Mikolaj Rybinski <mikolaj.rybinski@id.ethz.ch>
Date: Fri, 3 May 2019 10:43:15 +0200
Subject: [PATCH] styling update + minor misc text
---
05_classifiers_overview.ipynb | 15 ++++++++++++---
1 file changed, 12 insertions(+), 3 deletions(-)
diff --git a/05_classifiers_overview.ipynb b/05_classifiers_overview.ipynb
index 6957d36..cfcc4e8 100644
--- a/05_classifiers_overview.ipynb
+++ b/05_classifiers_overview.ipynb
@@ -120,6 +120,7 @@
"%config InlineBackend.figure_format = 'retina'\n",
"import warnings\n",
"warnings.filterwarnings('ignore', category=FutureWarning)\n",
+ "warnings.filterwarnings = lambda *a, **kw: None\n",
"from IPython.core.display import HTML; HTML(open(\"custom.html\", \"r\").read())"
]
},
@@ -1662,10 +1663,10 @@
"K(x, y) = e^{-\\gamma ||x - y||}\n",
"$$\n",
"\n",
- "It is a Gaussian-shaped similarity measure that returns `1` for the same points and declines exponentially to `0` with distance increasing between points.\n",
+ "It is a Gaussian-shaped similarity measure that returns `1` for the same points and declines exponentially to `0` with distance increasing between points, with a rate controlled by $\\gamma$ parameter.\n",
"\n",
"\n",
- "Using the so called **kernel trick**, SVC uses such similarity measure (kernel) as if applying mapping $\\phi$ without actually applying it, followed by a linear SVM.\n",
+ "Using the so called **kernel trick**, SVC uses such similarity measure (kernel) as if applying mapping $\\phi$, without actually applying it, followed by a linear SVM.\n",
"\n",
"\n",
"<table>\n",
@@ -1750,7 +1751,7 @@
"source": [
"## Exercise section\n",
"\n",
- "Play with different kernels and different gamma parameters of the `SVC` classifier. Which built-in kernels do work? Which gamma value to pick?"
+ "Play with different valuse of `kernel` and `gamma` parameters of the `SVC` classifier. Which built-in kernels do work? Which gamma value to pick?"
]
},
{
@@ -1886,6 +1887,14 @@
"source": [
"## Decision trees\n",
"\n",
+ "Let's explain what decision tree by looking at a decision tree for an email spam classification problem. It used 57 features, such as percentages of specific words or characters, and uninterrupted capital letters average, max and total lengths.\n",
+ "\n",
+ "<table>\n",
+ " <tr><td><img src=\"decision_tree-spam.png\" width=600px></td></tr>\n",
+ " <tr><td><center><sub>Source: Hastie, T., Tibshirani, R. and Friedman, J. H. (2009), <em>The elements of statistical learning data mining, inference, and prediction</em>.</sub></center></td></tr>\n",
+ "</table>\n",
+ "\n",
+ "\n",
"**TODO**\n",
"\n",
"- simple example incl. plot\n",
--
GitLab