Tech Companies and Academics Collaborating on COVID-19 Research

ARTIFICIAL INTELLIGENCE AI Powered Contact Center Messaging and AI-Open Source The collaboration between tech giants and academics on COVID-19 is a welcome move as the pandemic continues to ravage people across the globe. From Italy, to Spain and the United States, COVID-19 is increasing with hospitals under strain because of excess patients. Businesses are announcing … Read more

50 Deep Learning Interview Questions

Practice Problems and Solutions on Deep Learning Below are 25 questions on deep learning which can help you test your knowledge, as well as being a good review resource for interview preparation. 1. Why is it necessary to introduce non-linearities in a neural network? Solution: otherwise, we would have a composition of linear functions, which … Read more

Deploying a Deep Learning Model using Flask

Finding a framework for web deployment of a deep learning model I am creating the web deployment for a book I am writing for Manning Publications on deep learning with structured data. The audience for this book is interested in how to deploy a simple deep learning model. They need a deployment example that is … Read more

Three Data Science Technologies to Explore while you Self-Isolate: What are Docker, Airflow and…

Like in many other states (and even countries), Minnesotans were issued orders to stay inside to help flatten the COVID-19 infection rate curve. Besides giving my dog lots of walks, to pass the time as I stay home for the next few weeks I am prepared with several streaming services, Lego, puzzles, video games, and … Read more

Image Denoising with Gibbs Sampling (MCMC)  — Concepts and Code Implementation

There are three sections in this tutorial: Theoretical Concepts of MCMC and Gibbs Sampling Mathematical Deduction for the Problem Code Implementation in Python Markov chain Monte Carlo (MCMC) is a sampling method used to approximate the posterior distribution of a parameter of interest by randomly sampling from a probability distribution and constructing a Markov chain. … Read more

Death and Data Science

Why we need to think critically of data usage in the age of COVID-19 Photo by Nahel Abdul Hadi on Unsplash A little over a month ago, I wrote this article about our fear of death, and how tech companies will manipulate that fear to convince us to give up our privacy. As noted in … Read more

The Energy of the Vacuum

In this article, I will describe some significant consequences of the quantum energy of the vacuum. The latter exists in the background throughout the entire Universe. More specifically, I will explain the so-called Casimir effect in quantum field theory (QFT). The Casimir effect is a small attractive force that acts between two close parallel uncharged … Read more

Predicting Demand During a Crisis

Consumer demand is fundamentally different in the lockdown world: historical data and modeling infrastructure built on that data are no longer representative of the world we live in today. Certain product categories will feel the effect of this shift more than others, but we can assume that most product-store observations before February 2020 are biased. … Read more

Meet {tidycovid19}: Yet another Covid-19 related R Package

I have decided that the world needs another Covid-19 related R package. Not sure whether you agree, but the new package facilitates the direct download of various Covid-19 related data (including data on governmental measures) directly from authoritative sources. It also provides a flexible function and accompanying shiny app to visualize the spreading of the … Read more

Categories R Tags ExcerptFavorite

Installing spatial R packages on Ubuntu

This post explains how to quickly get key R packages for geographic research installed on Ubuntu, a popular Linux distribution. A recent thread on the r-spatial GitHub organization alludes to many considerations when choosing a Linux set-up for work with geographic data, ranging from the choice of Linux distribution (distro) to the use of binary … Read more

Categories R Tags ExcerptFavorite

Forget the hassles of Anchor boxes with FCOS: Fully Convolutional One-Stage Object Detection

This article is a detailed explanation of a new object detection technique proposed in the paper FCOS: Fully Convolutional One-Stage Object Detection published at ICCV’19. I decided to summarize this paper because it proposes a really intuitive and simple technique that solves the object detection problem. Stick around to know how it works. Anchor-Based Detectors … Read more

Visualizing the Coronavirus Pandemic with Choropleth Maps

The code to create this is as follows: # Import librariesimport numpy as np import pandas as pd import plotly as pyimport as pximport plotly.graph_objs as gofrom plotly.subplots import make_subplotsfrom plotly.offline import download_plotlyjs, init_notebook_mode, plot, iplotinit_notebook_mode(connected=True)# Read Datadf = pd.read_csv(“../input/novel-corona-virus-2019-dataset/covid_19_data.csv”)# Rename columnsdf = df.rename(columns={‘Country/Region’:’Country’})df = df.rename(columns={‘ObservationDate’:’Date’})# Manipulate Dataframedf_countries = df.groupby([‘Country’, ‘Date’]).sum().reset_index().sort_values(‘Date’, ascending=False)df_countries = df_countries.drop_duplicates(subset … Read more

Which Machine Learning Model Are You?

A lighthearted, quippy, and un-scientific description of how some classic models work Perhaps you have (against your better judgement) taken a quiz online about which of your favorite TV show characters you resemble, how old you seem based on your breakfast habits, or my personal favorite, which type of chicken nugget you are. These quizzes … Read more

Strategies for Model Debugging

Systematic ways to test and fix machine learning models Contents: Introduction1.1 A Quick Note on Trust and Understanding1.2 Example Problem and Dataset Detection Strategies2.1 Sensitivity Analysis2.2 Residual Analysis2.3 Benchmark Models2.4 Security Audits for ML Attacks Remediation Strategies3.1 Data Augmentation3.2 Noise Injection and Strong Regularization3.3 Model Editing3.4 Model Assertions3.5 Unwanted Sociological Bias Remediation3.6 Model Management and … Read more

Stylistic differences between R and Python in modelling data through neural networks

Neural networks in data science represent an attempt to reproduce a non-linear learning that occurs in the network of neurons found in nature. A neuron consists of dendrites to gather inputs from other neurons and combines the input information in order to generate a response when some threshold is reached which is sent to other … Read more

Building a Personalized Real-Time Fashion Collection Recommender

Traditionally, we represent images as a massive array of integers (3D array for RGB images and 1D array for grayscale images). These arrays are unwieldy and grow exponentially — we will need to keep track of millions of numbers to analyze hundreds of high-resolution images! It will be impossible to scale any modeling using arrays … Read more

Using Data Science for Customer Acquisition

Customer Segmentation Report and Customer Response prediction With the aim of bringing new customers to their brands, companies have been developing their and customer acquisition processes and strategies. It’s not easy though! The first step of any basic customer acquisition plan is to identify quality potential customers and, guess what! Data Science can help us … Read more

wrapr 2.0.0 up on CRAN

[This article was first published on R – Win-Vector Blog, and kindly contributed to R-bloggers]. (You can report issue about the content on this page here) Want to share your content on R-bloggers? click here if you have a blog, or here if you don’t. wrapr 2.0.0 is now up on CRAN. This means the … Read more

Categories R Tags ExcerptFavorite

Anti-coronavirus natural products and In silico screening

There is an urgent need to discover novel antivirals Viruses are responsible for a number of human pathologies including cancer, while complex syndromes like Alzheimer’s disease, type 1 diabetes and hepatocellular carcinoma have been also associated with viral infections. Moreover, due to increased global travel and rapid urbanization, epidemic outbreaks caused by emerging and re‐emerging … Read more

COVID-CXR: An open source explainable deep CNN model for predicting the presence of COVID-19 in…

Figure 1: An example of an explanation of the binary model’s prediction on a COVID-19 example in the test set. Colorized regions contributing toward (green) and against (red) prediction of COVID-19. Note from the editors: Towards Data Science is a Medium publication primarily based on the study of data science and machine learning. We are … Read more

Walkthrough: Mapping GIS Data in Python

It seems geographic data has never been more important than at this moment in history. This article will improve your understanding of geospatial information, allowing you an entry point to the rich world of geographic information science (GIS) through neat, easy-to-work with pandas DataFrames. Introduction | Activity | Resources By following this code-along, you’ll learn … Read more

Text Ming Comments of “A Plan to Get America Back to Work

Economy, public health, coronavirus, NLP Was intrigued by one of Thomas L. Friedman’s OpEd on New York Times last week: “A Plan to Get America Back to Work”. It advocated a Data-Driven approach to the COVID-19 Pandemic: that is, limiting the number of infections and deaths from the coronavirus, in the same time maximizing the … Read more

Building a Bot That Plays Videos for My Toddler

In order to complete the visual presence module, the next thing I needed to build was a fidget animation system which is based on a simple keyframe animation. A keyframe animation lets you animate an object such as a head by supplying its initial and final transform (position, scale and rotation) and the duration of … Read more

DATE/TIME Functions in SQL

Next, we can look at pulling a specific format out of a timestamp. The goal is to extract a portion out of a timestamp. For example, if we want just the month from the date 12/10/2018, we would get December (12). Let’s take a look at EXTRACT syntax EXTRACT(part FROM date) We state the type … Read more

Are Stock Returns Normally Distributed?

Since 1950, the average annual return of the S&P 500 has been approximately 8% and the standard deviation of that return has been 12%. I want to look at monthly returns so let’s translate these to monthly: Monthly Expected Return = 8%/12 = 0.66%Monthly Standard Deviation = 12%/(12^0.5) = 3.50% Let’s overlay the actual returns … Read more

Regular Expressions in Python

Regular expressions are special sequences of characters that define search patterns in texts. In this post, we will discuss how to use one of the most basic regular expression methods in python, ‘re.findall()’, to extract the beginning of string expressions. Let’s get started! To start, let’s import the python regular expression module, ‘re’: import re … Read more

A Newspaper for COVID-19 — The CoronaTimes

The most important thing while creating this Dashboard was acquiring the data. I am using two data sources: 1. Data from the European Centre for Disease Prevention and Control. The downloadable data file is updated daily and contains the latest available public data on COVID-19. Here is a snapshot of this data. def get_data(date):os.system(“rm cases.csv”)url … Read more

R Tutorial: Analyzing COVID-19 Data

Introduction to using R in the real world Source This semester at Yale I am taking an Introduction to Econometrics class where we utilize R and statistics in order to analyze various datasets. So, while I am socially distancing myself at home, I thought it would be interesting to apply some of the techniques that … Read more

Job Board Scraping with Rails

Build a scheduled job scraper with Ruby on Rails and Heroku Photo by Donald Giannatti on Unsplash Job-related data is one of my favorite to play with. It’s fun to parse requirements and analyze changes across time, companies and geography. While there are some great public databases with job-related information, it’s also pretty available to … Read more

Why testing completely skews Coronavirus case fatality rates

Ramping up testing is imperative. During the course of an infectious disease outbreak such as the COVID-19 pandemic the world is currently experiencing, it is crucial to understand the lethality of the illness— people want to know their likelihood of dying if they catch the disease. Multiple different measures are used to estimate this, and … Read more

A GARCH Tutorial in R

[This article was first published on R on msperlin, and kindly contributed to R-bloggers]. (You can report issue about the content on this page here) Want to share your content on R-bloggers? click here if you have a blog, or here if you don’t. Myself, Mauro Mastella, Daniel Vancin and Henrique Ramos, just finished a … Read more

Categories R Tags ExcerptFavorite

Whats Cooking ??

[This article was first published on Stencilled, and kindly contributed to R-bloggers]. (You can report issue about the content on this page here) Want to share your content on R-bloggers? click here if you have a blog, or here if you don’t. This R shiny app is for all the home chefs out there looking … Read more

Categories R Tags ExcerptFavorite

5 Lessons from Life as a Data Scientist

It’s been about a year since I started my first full-time role out of college, working as a data scientist at Squarespace. Recently, I decided to change things up and join a fast-growing startup called Hugo to work on all things Growth. This detour away from further specialization in data science surprised a lot of … Read more

Neural Style Transfer With TensorFlow Hub

Here comes the good bit. We will create an image with the content and style of the images below. Left: Content Image (Photo by Štefan Štefančík on Unsplash), Right: Style Image (Photo by adrianna geo on Unsplash) In order to successfully implement the process of Neural Style Transfer using two reference images, we’ll be leveraging … Read more

Are you using Python with APIs? Learn how to use a retry decorator!

Functions are first-class objects In Python, functions are first-class objects. A function is just like any other object. This fact, among other things, means that a function can be dynamically created, passed to a function itself, and even changed. Take a look at the following (albeit silly) example: def my_function(x):print(x)IN:my_function(2)OUT:2IN:my_function.yolo = ‘you live only once’print(my_function.yolo)OUT:’you … Read more

Stream Learning in Energy IoT Systems

Image taken from with a license to use Creative Commons Zero — CC0. The prediction of electrical power produced in combined cycle power plants is a key challenge in the electrical power and energy systems field. This power production can vary depending on environmental variables, such as temperature, pressure, and humidity. Thus, the business … Read more

It’s time to get a PhD in Coronavirus

Depending on where you live, you’ve probably been suffering from Corona-related anxiety for the last few months, weeks, or days. And all the downplaying, the fake news, even the funny memes, tend to heighten the angst, at least for me. If there is one cure for all this uncertainty, then it’s knowledge. I wrote a … Read more

Update #2 on Microsoft cloud services continuity

Since last week’s update, the global health pandemic continues to impact every organization—large or small—their employees, and the customers they serve. Everyone is working tirelessly to support all our customers, especially critical health and safety organizations across the globe, with the cloud services needed to sustain their operations during this unprecedented time. Equally, we are … Read more

It’s Time to Brace for Impact — A look at COVID-19 data shows our current measures are not enough

The second table showing the growth rates are an attempt to capture how quickly the numbers are growing. Growth rate is the variable “r” in the basic exponential growth calculation A=A0ert. The values were calculated daily using previous day (A0) to current day’s (A) total cases. In the Italy and Hubei data, numbers above 0.2 … Read more

RProtoBuf 0.4.17: Robustified

[This article was first published on Thinking inside the box , and kindly contributed to R-bloggers]. (You can report issue about the content on this page here) Want to share your content on R-bloggers? click here if you have a blog, or here if you don’t. A new release 0.4.17 of RProtoBuf is now on … Read more

Categories R Tags ExcerptFavorite

Handling Big Volume of Well Log Data with a Boosted Time-Efficiency with Python

How Python has revolutionized the way we interact with big data in the petroleum industry and so should we know about it A slice of the earth. Figure by Ani-Mate/ This is a story of a geophysicist who has been already getting tired of handling the big volume of well log data with manual input … Read more

Modeling Logistic Growth

Nonlinear Least Squares Estimation of the Logistic Growth Function Using Scipy in Python — Using China’s Coronavirus data In a previous article, I have explained how to model the spread of the Coronavirus outbreak using Exponential Growth. It was limited to the first phase of the outbreak since the big limitation of Exponential Growth is … Read more

A deep dive into glmnet: predict.glmnet

I’m writing a series of posts on various function options of the glmnet function (from the package of the same name), hoping to give more detail and insight beyond R’s documentation. In this post, instead of looking at one of the function options of glmnet, we’ll look at the predict method for a glmnet object … Read more

Categories R Tags ExcerptFavorite