Remix Institute

Accelerate to data science

Blog

Business AI for SMB’s: How To Compete With Amazon Using RemixAutoML

Posted on September 16, 2019October 10, 2019 by Nick Gausling and Douglas Pestana by Nick Gausling and Douglas Pestana

The modern business world is captivated by data science and analytics. It seems like in almost every enterprise there is talk of needing to bulk up the data science and…

Why Machine Learning is more Practical than Econometrics in the Real World

Posted on August 18, 2019August 25, 2019 by Adrian Antico by Adrian Antico

Motivation I’ve read several studies and articles that claim Econometric models are still superior to machine learning when it comes to forecasting. In the article, “Statistical and Machine Learning forecasting methods:…

Build Thousands of Automated Demand Forecasts in 15 Minutes Using AutoCatBoostCARMA in R

Posted on June 24, 2019August 18, 2019 by Adrian Antico by Adrian Antico

Over 20% of Amazon’s North American retail revenue can be attributed to customers who first tried to buy the product at a local store but found it out-of-stock, according to…

Automate Your KPI Forecasts With Only 1 Line of R Code Using AutoTS

Posted on May 28, 2019May 28, 2019 by Douglas Pestana - @DougVegas by Douglas Pestana - @DougVegas

If you are having the following symptoms at your company when it comes to business KPI forecasting, then maybe you need to look at automated forecasting: Ugly Excel spreadsheets with…

Companies Are Demanding Model Interpretability. Here’s How To Do It Right.

Posted on April 11, 2019April 22, 2019 by Adrian Antico by Adrian Antico

There are plenty of methods out there for machine learning model interpretability with fancy names but what is lacking in all of them? ICE and Partial Dependence Plots in…

The Easiest Way to Create Thresholds And Improve Your Classification Model

Posted on April 11, 2019April 22, 2019 by Adrian Antico by Adrian Antico

The Importance of Using Thresholds to Convert Predicted Probabilities into Class Labels You’d be surprised at how many data scientists don’t know how to turn their probabilities into class labels.…