Against the Current

Yeniden Sanayi(siz)leşme?

ABD Başkanı Trump, 2 Nisan tarihli bir kararname ile Kanada ve Meksika dışındaki çok sayıda ticaret ortağının mallarına 9 Nisan’da geçerli olmak üzere hem %10 genel hem de…

Classification with Scikit-learn

In the previous two posts of this series on scikit-learn, we focused on a regression problem and trained linear regression, decision trees, bagging, random forests, and…

Tree-Based Regression Models with Scikit-learn

As we continue to explore the scikit-learn package, we now consider the following tree-based models: decision trees, bagging, random forests, and boosting. In this post, we…

Scikit-learn: Sampling, Feature Engineering, and Pipelines

I have published four posts on machine learning in R and illustrated how to use the great tidymodels package to implement the most common machine learning tasks…

Machine Learning with tidymodels – Part 4

In the previous three posts, we worked on a dataset where the response variable was quantitative/continuous (the sale price of a house). Oftentimes, the response variable is…

Machine Learning with tidymodels – Part 3

If you had reviewed the code in the previous post, you would have noticed that tuning a boosted trees model takes a lot of time (approximately 20 minutes on my machine).…

Milli Kütüphane Okuma Salonları İçin Bir Öneri

Ankara’daki Milli Kütüphane’nin kentin en faydalı kamu kurumlarından biri olduğu söylenebilir. Verdiği diğer önemli hizmetlerin yanında, yüksek tavanlı büyük okuma salonları…

Machine Learning with tidymodels – Part 2

In the previous post, we focused on some fundamental steps involved in machine learning exercises: data splitting and resampling, preprocessing and model specification…

Machine Learning with tidymodels – Part 1

tidymodels is a meta-package for machine learning and statistical analysis in R. It offers tools to greatly simplify the model training process. Moreover, it shares the same…

Decomposing Inflation

From the late 1960s through the early 1980s, inflation rose by at least two to three times in many countries due to preceding high growth, tight labor markets, and two oil…