diff --git a/_posts/2018-04-29-ROC-space-and-AUC.md b/_posts/2018-04-29-ROC-space-and-AUC.md
index 0c421f0..72809ab 100755
--- a/_posts/2018-04-29-ROC-space-and-AUC.md
+++ b/_posts/2018-04-29-ROC-space-and-AUC.md
@@ -7,6 +7,8 @@ tags: [AUC, false positive rate, ROC, true positive rate, type I error, type II
image: cost_roc_space.png
---
+# KEVIN WAS HERE
+
Before discussing ROC curves and AUC, let's fix some terminology around the confusion matrix:
* **Condition positive (negative):** real positive (negative) case in the data
@@ -70,4 +72,4 @@ Given any two classifiers in ROC space, we can interpolate on the line segment b
## References
* *Measuring classifier performance: a coherent alternative to the area under the ROC curve* by David J. Hand
-* *An introduction to ROC analysis* by Tom Fawcett
\ No newline at end of file
+* *An introduction to ROC analysis* by Tom Fawcett
diff --git a/_site/ROC-space-and-AUC.html b/_site/ROC-space-and-AUC.html
index ff47aa8..433c983 100644
--- a/_site/ROC-space-and-AUC.html
+++ b/_site/ROC-space-and-AUC.html
@@ -43,8 +43,8 @@
-
-
+
+
@@ -114,6 +114,8 @@
+
KEVIN WAS HERE
+
Before discussing ROC curves and AUC, let’s fix some terminology around the confusion matrix:
diff --git a/_site/feed.xml b/_site/feed.xml
index 60c2430..ac6050c 100644
--- a/_site/feed.xml
+++ b/_site/feed.xml
@@ -1,4 +1,4 @@
-Jekyll2019-09-29T11:38:15-07:00http://localhost:4000/feed.xmlstatsandstuffa blog on statistics and machine learningScott RoyWhat makes a better score distribution?2019-09-29T00:00:00-07:002019-09-29T00:00:00-07:00http://localhost:4000/what-makes-a-better-score-distribution<p>Suppose I train two binary classifiers on some data, and after examining the score distributions of each, I see the results below. Which score distribution is better? (And by extension, which classifier is better?)</p>
+Jekyll2019-10-09T12:51:33-07:00http://localhost:4000/feed.xmlstatsandstuffa blog on statistics and machine learningScott RoyWhat makes a better score distribution?2019-09-29T00:00:00-07:002019-09-29T00:00:00-07:00http://localhost:4000/what-makes-a-better-score-distribution<p>Suppose I train two binary classifiers on some data, and after examining the score distributions of each, I see the results below. Which score distribution is better? (And by extension, which classifier is better?)</p>
<table>
<tbody>
@@ -1456,7 +1456,9 @@ F &= \frac{\text{SS}_{\text{reg}} / k}{\text{RSS} / (n-k-1)} \\
<h2 id="references">References</h2>
<ul>
<li><em>Adaptive Rejection Sampling for Gibbs Sampling</em> by W. R. Gilks and P. Wild</li>
-</ul>Scott RoyThere is almost no difference between knowing a distribution’s density (and thus knowing its mean, variance, mode, or anything else about it) and being able to sample from the distribution. On the one hand, if we can sample from a distribution, we can estimate the density with a histogram or kernel density estimator. Conversely, I’ll discuss ways to sample from a density in this post.ROC space and AUC2018-04-29T00:00:00-07:002018-04-29T00:00:00-07:00http://localhost:4000/ROC-space-and-AUC<p>Before discussing ROC curves and AUC, let’s fix some terminology around the confusion matrix:</p>
+</ul>Scott RoyThere is almost no difference between knowing a distribution’s density (and thus knowing its mean, variance, mode, or anything else about it) and being able to sample from the distribution. On the one hand, if we can sample from a distribution, we can estimate the density with a histogram or kernel density estimator. Conversely, I’ll discuss ways to sample from a density in this post.ROC space and AUC2018-04-29T00:00:00-07:002018-04-29T00:00:00-07:00http://localhost:4000/ROC-space-and-AUC<h1 id="kevin-was-here">KEVIN WAS HERE</h1>
+
+<p>Before discussing ROC curves and AUC, let’s fix some terminology around the confusion matrix:</p>
<ul>
<li><strong>Condition positive (negative):</strong> real positive (negative) case in the data</li>
@@ -1524,4 +1526,4 @@ Let <script type="math/tex">f_{+}(x)</script> be the score
<ul>
<li><em>Measuring classifier performance: a coherent alternative to the area under the ROC curve</em> by David J. Hand</li>
<li><em>An introduction to ROC analysis</em> by Tom Fawcett</li>
-</ul>Scott RoyBefore discussing ROC curves and AUC, let’s fix some terminology around the confusion matrix:
\ No newline at end of file
+</ul>Scott RoyKEVIN WAS HERE
\ No newline at end of file
diff --git a/_site/page2/index.html b/_site/page2/index.html
index a261e69..005c184 100644
--- a/_site/page2/index.html
+++ b/_site/page2/index.html
@@ -214,12 +214,14 @@
- Before discussing ROC curves and AUC, let’s fix some terminology around the confusion matrix:
+ KEVIN WAS HERE
+
+Before discussing ROC curves and AUC, let’s fix some terminology around the confusion matrix:
Condition positive (negative): real positive (negative) case in the data
True positive (negative): condition positive (negative) that is classified as positive (negative)
- False positive (negative): condition negative (positive) that is classifie... Read more
+ False positive (negative): condition negative (positive) t... Read more 29 Apr 2018 -
diff --git a/_site/rss-feed.xml b/_site/rss-feed.xml
index 1e24f4b..476a37d 100644
--- a/_site/rss-feed.xml
+++ b/_site/rss-feed.xml
@@ -149,7 +149,7 @@ There are many ways we could interpret “most related”:</p>
Scott Roy
- <p>Before discussing ROC curves and AUC, let’s fix some terminology around the confusion matrix:</p>
+ <h1 id="kevin-was-here">KEVIN WAS HERE</h1>
diff --git a/_site/sitemap.xml b/_site/sitemap.xml
index deb8cc2..282e9a6 100644
--- a/_site/sitemap.xml
+++ b/_site/sitemap.xml
@@ -103,10 +103,10 @@
http://localhost:4000/assets/img/backprop_params.pdf
-2019-09-12T00:31:53-07:00
+2019-10-09T12:47:33-07:00http://localhost:4000/assets/img/backprop_prevoutput.pdf
-2019-09-12T00:47:56-07:00
+2019-10-09T12:47:33-07:00
+