Improve your regression model's accuracy and predictability by uncovering hidden errors with these essential plots.

Sometimes input for document processing tasks such as OCR, table detection or text segmentation can be scanned or photo taken from hand that do not have ideal perspective - is rotated or spatially distorted in some way (warped document). If you are looking for my recommendations go straight to the last section of this article "Summary and recommendations".

This post describes techniques that help to get most accurate suggestions from the GitHub Copilot "Your AI pair programmer". For those who never heard of Copilot there is short introduction, if you already know Copilot - you can jump directly to section 4 - "How to get most of GitHub Copilot".

Learn about micro and macro averages in multiclass multilabel problems, the difference between multiclass and multilabel problems and when to use micro and macro averages.

Learn how to set up VSCode as a Nim IDE from scratch. Includes instructions for downloading VSCode, installing the Nim extension, configuring settings, and debugging Nim code.

Explore the most popular Zsh plugins from the 1800+ options on the Awesome Zsh plugins GitHub project. See which ones have the highest number of stars from the Zsh community.

Want to create beautiful visualizations from complex data? Discover the power of T-SNE for dimensionality reduction in Python.

Statistics can be tricky, but understanding kurtosis is a must for anyone who wants to avoid making common mistakes in statistical analyses. Learn how to interpret it in this comprehensive guide.

Explore methods to detect & fix errors in data, including validation, visualizations, statistical tests, cleaning techniques, machine learning & data quality tools. Get concise, easy to understand information with examples & links to external resources.

Overview of the available tools and methods for schema validation in pandas, examplary code snippets and recommendation for when to use given tool.

Learn about the evaluation of interpretability in machine learning with this guide. Discover different levels and methods for assessing the explainability of models.

Learn about metrics used to compare histograms with examples of how to calculate them in python. From Chi-Squared distance to Kullback-Leibler divergence and Earth Mover's distance. A comprehensive guide.

Using Plotly, Bokeh and Altair for interactive visualizations in the blog posts.

Description of GNU utils and other less standard tools that helps with processing data from CLI or with shell scripts.

Using puzzle on tracing the high profile spy as excuse to showcase Markov Chains and demonstrate usage and properties e.g. Stationary distribution

On the GitHub project Awesome Zsh plugins you can find 1700+ links to plugins, themes, and Zsh plugin managers/frameworks. The number of tools listed on that page is high and it is difficult to get orientation on which plugins gained already a good reputation from the Zsh users community. This post aims at identifying the most popular tools where popularity is measured by the number of stars that Github users added to a given plugin or tool.

This post proposes a 4-step path for learning Bayesian methods. The first step is going through the book "Bayesian methods for hackers", second, using complementary books for probability and statistics, the third, reading How to become a Bayesian in eight easy steps, and last, going through the book full of exercises - "Think Bayes".

"This post describe evaluation metrics used in Kaggle competitions where problem to solve is has regression nature. Eight different metrics are described, namely - Absolute Error (AE), Mean Absolute Error (MAE), Weighted Mean Absolute Error (WMAE), Pearson Correlation Coefficient, Spearman\u2019s Rank Correlation, Root Mean Squared Error (RMSE), Root Mean Squared Logarithmic Error (RMSLE), Mean Columnwise Root Mean Squared Error (MCRMSE)."

Guide on how to install TensorFlow cpu-only version - the case for machines without GPU supporting CUDA. Step-by-step procedure starting from creating conda environment till testing if TensorFlow and Keras Works.

Can the evolutionary approach crash the problem that brute-forcing will last far more than the age of the universe? This post shows how to attack the Traveling Salesman Problem using Darwin's approach. I'm describing the evolution model and design decisions. See the animation of how the population was evolving through the epochs.

Having several notebook-based projects behind you might result in a mess in the projects directory. Organize your Data Science project based on Jupyter notebooks in a way that one can navigate through it. Especially that "the one" will be most probably you in a few months time. To achieve that, keep your projects directory clean, name the project in a descriptive way and take care of the internal structure of the project.

Exploratory Data Analysis of the Kaggle's "What's cooking" competition dataset to get understanding what kind of data we are dealing with and get intuition of existing dependencies.

Python implementation of blockchain in few lines of code.

There is an exhaustive but curated list of Zsh plugins posted on GitHub project Awesome Zsh plugins. You can find there 800+ links to plugins, themes and Zsh plugin managers/frameworks. Even though it is a collection of awesome stuff the number is a bit high get orientation which plugins gained already good reputation from Zsh users community. In this post I will identify most popular plugins - those which have the highest number of stars.