Eric Maugendre<p><span class="h-card" translate="no"><a href="https://a.gup.pe/u/data" class="u-url mention" rel="nofollow noopener" target="_blank">@<span>data</span></a></span> <span class="h-card" translate="no"><a href="https://a.gup.pe/u/datadon" class="u-url mention" rel="nofollow noopener" target="_blank">@<span>datadon</span></a></span> 🧵</p><p>Redressing <a href="https://hachyderm.io/tags/Bias" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Bias</span></a>: "Correlation Constraints for Regression Models":<br>Treder et al (2021) <a href="https://doi.org/10.3389/fpsyt.2021.615754" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://</span><span class="ellipsis">doi.org/10.3389/fpsyt.2021.615</span><span class="invisible">754</span></a></p><p><a href="https://hachyderm.io/tags/dataDev" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>dataDev</span></a> <a href="https://hachyderm.io/tags/linearRegression" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>linearRegression</span></a> <a href="https://hachyderm.io/tags/modeling" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>modeling</span></a> <a href="https://hachyderm.io/tags/probability" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>probability</span></a> <a href="https://hachyderm.io/tags/probabilities" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>probabilities</span></a> <a href="https://hachyderm.io/tags/statistics" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>statistics</span></a> <a href="https://hachyderm.io/tags/stats" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>stats</span></a> <a href="https://hachyderm.io/tags/modelling" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>modelling</span></a> <a href="https://hachyderm.io/tags/regression" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>regression</span></a> <a href="https://hachyderm.io/tags/correctionRatio" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>correctionRatio</span></a> <a href="https://hachyderm.io/tags/skLearn" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>skLearn</span></a> <a href="https://hachyderm.io/tags/scikitLearn" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>scikitLearn</span></a> <a href="https://hachyderm.io/tags/python" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>python</span></a> <a href="https://hachyderm.io/tags/AIDev" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>AIDev</span></a></p>
sigmoid.social is one of the many independent Mastodon servers you can use to participate in the fediverse.
A social space for people researching, working with, or just interested in AI!
Administered by:
Server stats:
597active users
sigmoid.social: About · Profiles directory · Privacy policy
Mastodon: About · Get the app · Keyboard shortcuts · View source code · v4.4.3
#Sklearn
0 posts · 0 participants · 0 posts today
Eric Maugendre<p>"Feature importance helps in understanding which features contribute most to the prediction"</p><p>A few lines with <a href="https://hachyderm.io/tags/sklearn" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>sklearn</span></a>: <a href="https://mljourney.com/sklearn-linear-regression-feature-importance/" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://</span><span class="ellipsis">mljourney.com/sklearn-linear-r</span><span class="invisible">egression-feature-importance/</span></a> </p><p><a href="https://hachyderm.io/tags/interpretability" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>interpretability</span></a> <a href="https://hachyderm.io/tags/explainability" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>explainability</span></a> <a href="https://hachyderm.io/tags/AIethics" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>AIethics</span></a> <a href="https://hachyderm.io/tags/compliance" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>compliance</span></a> <a href="https://hachyderm.io/tags/taxonomy" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>taxonomy</span></a> <a href="https://hachyderm.io/tags/ethicalAI" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ethicalAI</span></a> <a href="https://hachyderm.io/tags/AIevaluation" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>AIevaluation</span></a> <a href="https://hachyderm.io/tags/linearRegression" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>linearRegression</span></a> <a href="https://hachyderm.io/tags/featureEngineering" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>featureEngineering</span></a></p>
Eric Maugendre<p><span class="h-card" translate="no"><a href="https://a.gup.pe/u/datadon" class="u-url mention" rel="nofollow noopener" target="_blank">@<span>datadon</span></a></span></p><p><a href="https://hachyderm.io/tags/Lasso" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Lasso</span></a> <a href="https://hachyderm.io/tags/LinearRegression" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>LinearRegression</span></a> "is useful in some contexts due to its tendency to prefer solutions with fewer non-zero coefficients, effectively reducing the number of features upon which the given solution is dependent"</p><p><a href="https://scikit-learn.org/stable/modules/linear_model.html#lasso" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://</span><span class="ellipsis">scikit-learn.org/stable/module</span><span class="invisible">s/linear_model.html#lasso</span></a> 🧵</p><p><a href="https://hachyderm.io/tags/dataDev" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>dataDev</span></a> <a href="https://hachyderm.io/tags/AIDev" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>AIDev</span></a> <a href="https://hachyderm.io/tags/ML" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ML</span></a> <a href="https://hachyderm.io/tags/sklearn" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>sklearn</span></a> <a href="https://hachyderm.io/tags/python" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>python</span></a> <a href="https://hachyderm.io/tags/interpretability" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>interpretability</span></a></p>
Peter Drake<p>I'm playing with the California Housing dataset built into sklearn.</p><p>One census block group has an average number of bedrooms per household of 0.83 and an average number of household members of 1243.</p><p>Huh?</p><p><a href="https://mstdn.social/tags/DataScience" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>DataScience</span></a> <a href="https://mstdn.social/tags/python" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>python</span></a> <a href="https://mstdn.social/tags/sklearn" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>sklearn</span></a></p>
Peter Jachim<p>I just did my first project using the <a href="https://mastodon.world/tags/mlflow" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>mlflow</span></a> library to track metrics on iterations of manual tuning of an <a href="https://mastodon.world/tags/sklearn" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>sklearn</span></a> pipeline, it works great and gives me some idea of the search space before moving into automated hyperparameter tuning.</p><p>I am using it in a super basic way, as an alternative to creating a gazillion cells with comments tracking metrics, does anyone have any favorite features to check out for taking mlflow to the next level?<br><a href="https://mastodon.world/tags/machinelearning" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>machinelearning</span></a> <a href="https://mastodon.world/tags/python" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>python</span></a> <a href="https://mastodon.world/tags/MLOps" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>MLOps</span></a> <a href="https://mastodon.world/tags/scikitlearn" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>scikitlearn</span></a></p>
➴➴➴Æ🜔Ɲ.Ƈꭚ⍴𝔥єɼ👩🏻💻<p>I genuinely miss PyMC2. The <a href="https://lgbtqia.space/tags/PyMC" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>PyMC</span></a> and <a href="https://lgbtqia.space/tags/Arviz" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Arviz</span></a> APIs changes so frequently, that it's impossible to know what the standard approach to anything is.</p><p><a href="https://lgbtqia.space/tags/Bayesian" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Bayesian</span></a> <a href="https://lgbtqia.space/tags/Statistics" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Statistics</span></a> in <a href="https://lgbtqia.space/tags/Python" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Python</span></a> should be easy. </p><p>To be honest, I'd really like a well maintained <a href="https://lgbtqia.space/tags/SkLearn" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>SkLearn</span></a> module for it.</p>
Joxean Koret (@matalaz)<p>Uhm... if I get a decision tree like the one shown in the picture, does it mean that I only need the columns shown in the tree for training and validation, right? I would only need the columns 2 and 3 (x[2], x[3]), isn't it? Or am I missing something else?</p><p><a href="https://mastodon.social/tags/Sklearn" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Sklearn</span></a> <a href="https://mastodon.social/tags/MachineLearning" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>MachineLearning</span></a> <a href="https://mastodon.social/tags/ML" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ML</span></a> <a href="https://mastodon.social/tags/DecisionTree" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>DecisionTree</span></a></p>
IB Teguh TM<p><a href="https://mastodon.social/tags/LinearRegression" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>LinearRegression</span></a> <a href="https://mastodon.social/tags/Python" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Python</span></a> <a href="https://mastodon.social/tags/Sklearn" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Sklearn</span></a><br>Dive into predictive modeling with our comprehensive guide on linear regression using Python and sklearn. Learn step-by-step implementation, result interpretation, and data visualization techniques. Perfect for beginners</p><p><a href="https://teguhteja.id/mastering-linear-regression-with-python-and-sklearn-a-step-by-step-guide/" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://</span><span class="ellipsis">teguhteja.id/mastering-linear-</span><span class="invisible">regression-with-python-and-sklearn-a-step-by-step-guide/</span></a></p>
Joxean Koret (@matalaz)<p>When training a model it turns out that I get better results with a small dataset than with a bigger dataset. This is what is called overfiting, right?<br><a href="https://mastodon.social/tags/MachineLearning" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>MachineLearning</span></a> <a href="https://mastodon.social/tags/Sklearn" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Sklearn</span></a></p>
Joxean Koret (@matalaz)<p>Dear Machine Learning people: when a problem can be solved using both a regressor and a classifier, which method would you choose? Or you simply try both and then choose whatever worked better? Any rule or set of rules to try to determine which method should work better?</p><p><a href="https://mastodon.social/tags/MachineLearning" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>MachineLearning</span></a> <a href="https://mastodon.social/tags/sklearn" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>sklearn</span></a> <a href="https://mastodon.social/tags/ML" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ML</span></a> <a href="https://mastodon.social/tags/Diaphora" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Diaphora</span></a></p>
:rss: Qiita - 人気の記事<p>【実装】ボストンの住宅価格推測AIを作ろう【後編】<br><a href="https://qiita.com/realmadridmarcelo/items/0ac4490de2007e43f65c?utm_campaign=popular_items&utm_medium=feed&utm_source=popular_items" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://</span><span class="ellipsis">qiita.com/realmadridmarcelo/it</span><span class="invisible">ems/0ac4490de2007e43f65c?utm_campaign=popular_items&utm_medium=feed&utm_source=popular_items</span></a></p><p><a href="https://rss-mstdn.studiofreesia.com/tags/qiita" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>qiita</span></a> <a href="https://rss-mstdn.studiofreesia.com/tags/Python" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Python</span></a> <a href="https://rss-mstdn.studiofreesia.com/tags/%E6%A9%9F%E6%A2%B0%E5%AD%A6%E7%BF%92" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>機械学習</span></a> <a href="https://rss-mstdn.studiofreesia.com/tags/%E5%85%A5%E9%96%80" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>入門</span></a> <a href="https://rss-mstdn.studiofreesia.com/tags/pandas" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>pandas</span></a> <a href="https://rss-mstdn.studiofreesia.com/tags/sklearn" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>sklearn</span></a></p>
James Ashford<p>In my job as a data analyst, I come across many different types of problems to solve. Some are relatively easy to solve, others not so much. That was until recently, where I came across a problem I have never given much thought before. That was until now.</p><p>What is the problem? Finding multiple peaks in a dataset.</p><p>You might think, this sounds […]</p><p><a href="https://jrashford.com/2024/03/25/finding-peaks-in-a-dataset-and-why-it-is-not-straightforward/" class="" rel="nofollow noopener" target="_blank">https://jrashford.com/2024/03/25/finding-peaks-in-a-dataset-and-why-it-is-not-straightforward/</a></p>
synlogic<p>anyone know of a FOSS lib equiv to Python's Scikit-learn (sklearn) but in/for Go?</p><p>(and to forestall an obvious suggestion which is likely a non-starter for my needs: yes I am aware of idea of wrapping it or otherwise linking out to it from Go, that is my worst case fallback, but avoiding it. ideal is a 100% pure Go source-to-binary solution)</p><p><a href="https://toot.io/tags/Golang" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Golang</span></a><br><a href="https://toot.io/tags/Python" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Python</span></a><br><a href="https://toot.io/tags/sklearn" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>sklearn</span></a><br><a href="https://toot.io/tags/ScikitLearn" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ScikitLearn</span></a><br><a href="https://toot.io/tags/ML" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ML</span></a><br><a href="https://toot.io/tags/stats" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>stats</span></a><br><a href="https://toot.io/tags/statistics" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>statistics</span></a><br><a href="https://toot.io/tags/math" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>math</span></a><br><a href="https://toot.io/tags/FOSS" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>FOSS</span></a></p>
PyData Madrid<p>Ya está abierto el registro para nuestra reunión de febrero: 🔍 Eficiencia operacional con LLMs y pipelines de scikit-learn, este mes en las oficinas de Adyen</p><p><a href="https://www.meetup.com/pydata-madrid/events/299189759/" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://www.</span><span class="ellipsis">meetup.com/pydata-madrid/event</span><span class="invisible">s/299189759/</span></a></p><p>¡Nos vemos el jueves 22 a las 19:00! Y después, networking 🍻</p><p><a href="https://masto.ai/tags/PyDataMadrid" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>PyDataMadrid</span></a> <a href="https://masto.ai/tags/PyData" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>PyData</span></a> <a href="https://masto.ai/tags/python" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>python</span></a> <a href="https://masto.ai/tags/ai" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ai</span></a> <a href="https://masto.ai/tags/llm" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>llm</span></a> <a href="https://masto.ai/tags/scikitlearn" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>scikitlearn</span></a> <a href="https://masto.ai/tags/sklearn" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>sklearn</span></a></p>
FormaK<p><span class="h-card" translate="no"><a href="https://fosstodon.org/@buck" class="u-url mention" rel="nofollow noopener" target="_blank">@<span>buck</span></a></span> The feature has landed! FormaK now supports hyper-parameter selection and cross validation with a new structured state machine interface. Under the hood it’s using scikit-learn. As always, it can be built into a <a href="https://fosstodon.org/tags/Python" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Python</span></a> or <a href="https://fosstodon.org/tags/Cpp" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Cpp</span></a> model or <a href="https://fosstodon.org/tags/KalmanFilter" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>KalmanFilter</span></a></p><p><a href="https://github.com/buckbaskin/formak/pull/21" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://</span><span class="ellipsis">github.com/buckbaskin/formak/p</span><span class="invisible">ull/21</span></a></p><p><a href="https://fosstodon.org/tags/sklearn" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>sklearn</span></a> <a href="https://fosstodon.org/tags/OpenSource" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>OpenSource</span></a></p>
scikit-learn<p>Discover scikit-learn 1.4 and its:<br>🟢 5 major features & 13 features<br>🔵 14 efficiency improvements & 23 enhancements<br>🟡 15 API changes<br>🔴 38 fixes</p><p>More details in the changelog: <a href="https://bit.ly/3tWlZA3" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://</span><span class="">bit.ly/3tWlZA3</span><span class="invisible"></span></a><br>or in the release highlights: <a href="https://bit.ly/3Hsoddm" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://</span><span class="">bit.ly/3Hsoddm</span><span class="invisible"></span></a></p><p>You can upgrade with pip as usual: <br>pip install -U scikit-learn</p><p>Or using the conda-forge builds: <br>conda install -c conda-forge scikit-learn</p><p>Thanks again to all the +80 contributors! </p><p><a href="https://fosstodon.org/tags/scikitlearn" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>scikitlearn</span></a> <a href="https://fosstodon.org/tags/Python" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Python</span></a> <a href="https://fosstodon.org/tags/sklearn" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>sklearn</span></a> <a href="https://fosstodon.org/tags/datascience" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>datascience</span></a> <a href="https://fosstodon.org/tags/opensource" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>opensource</span></a></p>
Olivier Grisel<p>I ran a quick Gradient Boosted Trees vs Neural Nets check using scikit-learn's dev branch which makes it more convenient to work with tabular datasets with mixed numerical and categorical features data (e.g. the Adult Census dataset).</p><p>Let's start with the GBRT model. It's now possible to reproduce the SOTA number of this dataset in a few lines of code 2 s (CV included) on my laptop.</p><p>1/n</p><p><a href="https://sigmoid.social/tags/sklearn" class="mention hashtag" rel="tag">#<span>sklearn</span></a> <a href="https://sigmoid.social/tags/PyData" class="mention hashtag" rel="tag">#<span>PyData</span></a> <a href="https://sigmoid.social/tags/MachineLearning" class="mention hashtag" rel="tag">#<span>MachineLearning</span></a> <a href="https://sigmoid.social/tags/TabularData" class="mention hashtag" rel="tag">#<span>TabularData</span></a> <a href="https://sigmoid.social/tags/GradientBoosting" class="mention hashtag" rel="tag">#<span>GradientBoosting</span></a> <a href="https://sigmoid.social/tags/DeepLearning" class="mention hashtag" rel="tag">#<span>DeepLearning</span></a> <a href="https://sigmoid.social/tags/Python" class="mention hashtag" rel="tag">#<span>Python</span></a></p>
Buck Baskin<p>I’m starting a new feature for <span class="h-card" translate="no"><a href="https://fosstodon.org/@formak" class="u-url mention" rel="nofollow noopener" target="_blank">@<span>formak</span></a></span>: semi-automated hyper-parameter selection for models and Kalman Filters.</p><p>You can read the design doc for the feature here: <a href="https://github.com/buckbaskin/formak/blob/hyperparameter-selection/docs/designs/hyperparameter_selection.md" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://</span><span class="ellipsis">github.com/buckbaskin/formak/b</span><span class="invisible">lob/hyperparameter-selection/docs/designs/hyperparameter_selection.md</span></a></p><p>Feedback on the design is welcome here or on GitHub</p><p><a href="https://fosstodon.org/tags/Python" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Python</span></a> <a href="https://fosstodon.org/tags/sklearn" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>sklearn</span></a> <a href="https://fosstodon.org/tags/KalmanFilter" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>KalmanFilter</span></a> <a href="https://fosstodon.org/tags/OpenSource" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>OpenSource</span></a> <a href="https://fosstodon.org/tags/design" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>design</span></a></p>
Olivier Grisel<p>scikit-learn 1.3.1 is out!</p><p>This release fixes a bunch of annoying bugs. Here is the changelog:</p><p><a href="https://scikit-learn.org/stable/whats_new/v1.3.html#version-1-3-1" target="_blank" rel="nofollow noopener" translate="no"><span class="invisible">https://</span><span class="ellipsis">scikit-learn.org/stable/whats_</span><span class="invisible">new/v1.3.html#version-1-3-1</span></a></p><p>Thanks very much to all bug reporters, PR authors and reviewers and thanks in particular to <span class="h-card" translate="no"><a href="https://fosstodon.org/@glemaitre" class="u-url mention">@<span>glemaitre</span></a></span>, the release manager of 1.3.1.</p><p><a href="https://sigmoid.social/tags/PyData" class="mention hashtag" rel="tag">#<span>PyData</span></a> <a href="https://sigmoid.social/tags/SciPy" class="mention hashtag" rel="tag">#<span>SciPy</span></a> <a href="https://sigmoid.social/tags/sklearn" class="mention hashtag" rel="tag">#<span>sklearn</span></a> <a href="https://sigmoid.social/tags/Python" class="mention hashtag" rel="tag">#<span>Python</span></a> <a href="https://sigmoid.social/tags/machinelearning" class="mention hashtag" rel="tag">#<span>machinelearning</span></a></p>
Olivier Grisel<p>🧵 I recently dived into a rabbit hole when attempting to fix the tests for <a href="https://sigmoid.social/tags/sklearn" class="mention hashtag" rel="tag">#<span>sklearn</span></a>'s OLS and Ridge regression solvers.</p><p>On the theoretical side, I now understand that the minimum norm solution for the centered problem without intercept is also the minimum norm solution for the original problem (with intercept). Ridge/OLS on centered X & y followed by intercept computation is the approach (hereafter name type "a") we have been using for years.</p><p><a href="https://raw.githubusercontent.com/ogrisel/minimum-norm-ols/main/minimum-norm-ols-intercept.pdf" target="_blank" rel="nofollow noopener" translate="no"><span class="invisible">https://</span><span class="ellipsis">raw.githubusercontent.com/ogri</span><span class="invisible">sel/minimum-norm-ols/main/minimum-norm-ols-intercept.pdf</span></a></p><p><a href="https://sigmoid.social/tags/PyData" class="mention hashtag" rel="tag">#<span>PyData</span></a> <a href="https://sigmoid.social/tags/Statistics" class="mention hashtag" rel="tag">#<span>Statistics</span></a></p>
TrendingLive feeds
Mastodon is the best way to keep up with what's happening.
Follow anyone across the fediverse and see it all in chronological order. No algorithms, ads, or clickbait in sight.
Create accountLoginDrag & drop to upload