## Time Series Imputation using Matrix Recovery

Missing values in time series are not straightforward to deal with. In my book on time series forecasting, Modern Time Series Forecasting with Python, I talk about a few techniques, right from thinking about the missing values in the right way to some algorithmic techniques to deal with missing data (like seasonal interpolation). Originally, that … Continue reading Time Series Imputation using Matrix Recovery

## Neural Networks – A Linear Algebra Perspective

Unlike my regular blog posts, this one is going to be a very short one - crisp and to the point. Deep Learning has been touted as the next big thing in data analytics and things have gotten so hyped that a lot of people (even practitioners) have started to consider it as magic. I'm … Continue reading Neural Networks – A Linear Algebra Perspective

## Mixture Density Networks: Probabilistic Regression for Uncertainty Estimation

Uncertainty is all around us. It is present in every decision we make, every action we take. And this is especially true in business decisions where we plan for the future. But in spite of that, all of our predictive models that we use in business ignore uncertainty. Suppose you are the manager of the … Continue reading Mixture Density Networks: Probabilistic Regression for Uncertainty Estimation

## Neural Oblivious Decision Ensembles(NODE) – A State-of-the-Art Deep Learning Algorithm for Tabular Data

Deep Learning brought about revolutions in many machine learning problems from the field of Computer Vision, Natural Language Processing, Reinforcement Learning, etc. But tabular data still remains firmly under classical machine learning algorithms, namely the gradient boosting algorithms(I have a whole series on different Gradient Boosting algorithms, if you are interested). Intuitively, this is strange, … Continue reading Neural Oblivious Decision Ensembles(NODE) – A State-of-the-Art Deep Learning Algorithm for Tabular Data

## PyTorch Tabular – A Framework for Deep Learning for Tabular Data

It is common knowledge that Gradient Boosting models, more often than not, kick the asses of every other machine learning models when it comes to Tabular Data. I have written extensively about Gradient Boosting, the theory behind and covered the different implementations like XGBoost, LightGBM, CatBoost, NGBoost etc. in detail. The unreasonable effectiveness of Deep … Continue reading PyTorch Tabular – A Framework for Deep Learning for Tabular Data

## How to Train and Deploy Custom AI-Generated Quotes using GPT2, FastAPI, and ReactJS

The Problem Good quotes help make us stronger. What is truly inspiring about quotes is not their tone or contentedness but how those who share them reflect life experiences that really serve others. I didn't write the above quote about quotes(Quote-ception), but an AI model I trained did. And it says it better than I … Continue reading How to Train and Deploy Custom AI-Generated Quotes using GPT2, FastAPI, and ReactJS

## Intermittent Demand Forecasting with Deep Renewal Processes

Let's face it, anyone who has worked on Time Series Forecasting problems in the retail, logistics, e-commerce etc. would have definitely cursed that long tail which never behaves. The dreaded intermittent time series which makes the job of a forecaster difficult. This nuisance renders most of the standard forecasting techniques impractical, raises questions about the … Continue reading Intermittent Demand Forecasting with Deep Renewal Processes

## Forecast Error Measures: Intermittent Demand

In the previous few blog posts, we've seen all the popular forecast measures used in practice. But all of them were really focused on smooth and steady time series. But there is a whole different breed of time series in real life - intermittent and lumpy demand. Source Casually, we call intermittent series as series … Continue reading Forecast Error Measures: Intermittent Demand

## Forecast Error Measures: Scaled, Relative, and other Errors

Following through from my previous blog about the standard Absolute, Squared and Percent Errors, let's take a look at the alternatives - Scaled, Relative and other Error measures for Time Series Forecasting. Taxonomy of Forecast Error Measures Both Scaled Error and Relative Error are extrinsic error measures. They depend on another reference forecast to evaluate … Continue reading Forecast Error Measures: Scaled, Relative, and other Errors

## Forecast Error Measures: Understanding them through experiments

Measurement is the first step that leads to control and eventually improvement.H. James Harrington In many business applications, the ability to plan ahead is paramount and in a majority of such scenario we use forecasts to help us plan ahead. For eg., If I run a retail store, how many boxes of that shampoo should … Continue reading Forecast Error Measures: Understanding them through experiments