Year: 2022

The recommended articles the author has read this week.
This letter is posted every Monday.

This week, the articles are about a Machine learning project environment, e. g. MLflow, and Jupyter Notebook.

In the environment of a machine learning project, it is very important to prepare not only a data analysis environment such as Jupyter Notebook but also an MLflow environment for managing experiment records. Managing experiment records helps to improve reproducibility and project promotion efficiency.

Containerize your whole Data Science Environment (or anything you want) with Docker-Compose

To build the environment of a machine learning project, docker-compose is a powerful tool. By docker-compose, we can build data-analysis and experiment-management experiments separately. This article tells us how to build such a style environment.

Manage your machine learning life cycle with MLflow in Python

This article shows one example of how to use the MLflow tracking server, which is a tool for managing experiment records. There are several styles to build the MLflow tracking server. This article suggests one of the helpful styles.

The recommended articles the author has read this week.
This letter is posted every Monday.

This week, the articles are about an explanative AI.

Top AutoML Python libraries in 2022

AutoML(Automated machine learning) is one of the hot topics. This article introduces the notable AutoML libraries, e.g. PyCaret, AtoKeras, and AutoGluon.

Data Drift Explainability: Interpretable Shift Detection with NannyML

NannyML is an open-source python library, a tool for estimating post-deployment model performance. Drift is one of the hot topics in MLOps, and this library is based on an interesting algorithm.

The recommended articles the author has read this week.
This letter is posted every Monday.

This week, the articles are about an explanative AI.

What Is a Transformer Model?

A transformer is a recent novel AI model which is based on a neural network. This model can learn context among sequence data such as sentences in a language. An attention mechanism in a transformer may be one of the most valuable knowledge we should learn.

Visualizing multicollinearity in Python

Multicollinearity is one of the annoying problems for a data scientist. In this article, we can learn how to visualize the relationships of multicollinearity between features.

The recommended articles the author has read this week.
This letter is posted every Monday.

This week, the articles are about an explanative AI.

Our approach to building transparent and explainable AI systems

An informative article whose topic is an explainable AI model. It is well known that it is difficult to build an explainable AI system. In this article, the author thinks, there is a hint.

FastTreeSHAP: Accelerating SHAP value computation for trees

One of the powerful methods to build an explainable AI model is to use a SHAP method. However, we face a computational problem when using a tree-based ML model with a large dataset. In this article, the new SHAP-based library is introduced. The new one is so faster.

The recommended articles the author has read this week.
This letter is posted every Monday.

This week, the articles about VAE(Variational Autoencoder) are introduced.

Pyro tutorial: Variational Autoencoders

A tutorial of VAE by Pyro, which is a universal probabilistic programming language (PPL) written in Python and supported by PyTorch on the backend. 

Variational Autoencoder: Intuition and Implementation

The theoretical background of VAE is written in detail.

The author also introduces several new features in the following articles.

The recommended articles the author has read this week.
This letter is posted every Monday.

Include diagrams in your Markdown files with Mermaid

An informative GitHub blog about how to create diagrams in markdown.

Practical Quantization in PyTorch

The PyTorch official blog about a quantization method. Quantization is a method for making your DNN run faster and with lower memory.

The author also introduces several new features in the following articles.

The recommended articles the author has read this week.
This letter is posted every Monday.

Comparison of AutoML solutions 2021

AutoML is one of the hot topics. Recently, however, there are many AutoML tools. In this article, we can check the recent trend.

Run Your Python Code as Quickly as C++

We sometimes face the problem that our python code is NOT as fast as we expect. This article introduces a really simple method to fast our python code.

The author also introduces several new features in the following articles.

The recommended articles the author has read this week.
This letter is posted every Monday.

T-distributed Stochastic Neighbor Embedding(t-SNE)

The techniques to reduce the dimensionality of a dataset are important because we can understand a dataset visually. Principal component analysis(PCA) is known as one of the most famous techniques. However, t-SNE is also might be a good choice because this technique can reflect the nonlinearity than that of PCA.

What makes LightGBM lightning fast?

LightGBM, one of the decision-tree-method libraries, is known as the faster library. This article demonstrates features of how to facilitate computational speed.

The author also introduces several new features in the following articles.