<div id="readability-page-1" class="page"><div><header>
<div>

<form role="search" method="get" action="http://mkseo.pe.kr/blog/"><label for="wp-block-search__input-1">Search</label></form></div>
</header>


<main>
<div>

<h2>Two interesting cross validation in scikit learn</h2>


<div>
<div><p><time datetime="2024-06-01T16:00:12+09:00">June 1, 2024</time></p></div>



<div>
<p>Tags:</p>


<div><p><a href="http://mkseo.pe.kr/blog/?tag=machine-learning" rel="tag">machine learning</a></p></div></div>
</div>
</div>


<div>
<p>Scikit is excellent esp when considering these advanced tools.</p>



<p>One is <a href="https://scikit-learn.org/stable/modules/calibration.html">calibrated classifier cv</a>. It tries to match model’s probability with the actually observed probability. </p>



<p>The other is <a href="https://scikit-learn.org/stable/auto_examples/release_highlights/plot_release_highlights_1_5_0.html#tunedthresholdclassifiercv-tuning-the-decision-threshold-of-a-binary-classifier">TunedThresholdClassifierCV</a> yet another interesting cv. If application requires different scores, e.g.&nbsp; F1,&nbsp; one can tune the decision threshold using it.</p>
</div>

<div>
		
		<h2>Comments</h2>
		

		

		

		

		<div id="respond">
		<h3 id="reply-title">Leave a Reply</h3><form action="http://mkseo.pe.kr/blog/wp-comments-post.php" method="post" id="commentform" novalidate=""><p><span id="email-notes">Your email address will not be published.</span> <span>Required fields are marked <span>*</span></span></p><p><label for="comment">Comment <span>*</span></label> </p><p><label for="author">Name</label> </p>
<p><label for="email">Email</label> </p>
<p><label for="url">Website</label> </p>
<p> <label for="wp-comment-cookies-consent">Save my name, email, and website in this browser for the next time I comment.</label></p>
</form>	</div>
	
	</div></main>


</div></div>

Passion is like genius; a miracle.

Scikit is excellent esp when considering these advanced tools. One is calibrated classifier cv. It tries to match model’s probability with the actually observed probability. The other is TunedThresholdClassifierCV yet another interesting cv. If application requires different scores, e.g.  F1,  one can tune the decision threshold using it.

Two interesting cross validation in scikit learn

댓글