Data Science Austria

Latest News about


How to tune hyperparameters of tSNE

Three simple rules to make beautiful tSNE plots Image source This is the second post of the column Mathematical Statistics and Machine Learning for Life Sciences. In the first post we discussed whether and where in Life Sciences we have Big Data suitable for Machine / Deep Learning, and emphasized

How to Choose Between Multiple Models

Key Concepts to Remember When Comparing Models In a previous article we discussed the concepts of underfitting and overfitting, how they can lead to models that don’t match the available data, how to identify each issue, and how to identify models that do fit the data well. These concepts can

A Recipe for using Open Source Machine Learning models

A step-by-step approach to finding, evaluating and using open source neural network models Photo by Luca Bravo on Unsplash Machine learning continues to produce state of the art (SOTA) results for an increasing variety of tasks and more companies are looking to ML to solve their problems. With the incredibly

The Hundred-Page Machine Learning Book Book Review

Chapter 9 & 10 — Learning without labels and other forms of learning Unsupervised learning is when your data doesn’t have labels. It’s a hard problem because you don’t have a ground truth to judge your model against. The book looks at two ways of dealing with unlabelled data, density

Generating Startup names with Markov Chains

Because generating new content doesn’t need to be complicated The most interesting applications of Machine Learning are, without a doubt, the generative models. The idea of finding patterns in data and generating new content that at the same time is similar to your data, but unique in its own way,

A Quant’s Intro to Portfolio Hedging

An intro to portfolio hedging, no finance background required… All analysis and programming efforts in this article can be found at: Poseyy/MarketAnalysis Portfolio Theory, Options Theory, & Quant Finance. Contribute to Poseyy/MarketAnalysis development by creating an… This article is purely introductory and experienced options traders will have to wait

Influencer Marketing using Web Scraping

Photo by Kaleidico on Unsplash Influencer marketing is not a foreign concept to us anymore. In the US retail business, there are only 33% of the retailers haven’t used or do not intend to use influencer marketing of any kind. However, it is a great way to lift your brand

The Easiest Python Numpy Tutorial Ever

Pie is delicious! Python is by far one of the easiest programming languages to use. Writing programs is intuitive and so is reading the code itself — it’s almost like plain English! One of Python’s greatest strengths is its endless supply of powerful libraries. Many of these libraries are written

Working with TFRecords and tf.train.Example In this tutorial, I will go over how to save and read data in TFRecord format which is the recommended format by Tensorflow. We are going to use the protocol buffer message tf.train.Example. I will do everything in Tensorflow 2.0 so let’s install it first. !pip install tensorflow==2.0.0-beta1import tensorflow

How to Ease the Pain of Working with Imbalanced Data

A summary of methods and resources for creating a model using an imbalanced dataset You’ve finally collected and cleaned your data and have even completed some exploratory data analysis (EDA). All that hard work has finally paid off — time to start playing with models! However, you quickly realize that