Blog

A data science blog for Petroleum Engineering.

Topics covered

Subject ▸ machine learning

A compilation of Machine Learning examples

21 November 2020

Introduction This is a compilation of machine learning examples that I found. They are easy to understand, they address a fundamental principle, they explain why they chose a particular algorithm. Some of them you will find very detailed; others are short and straight to the point. Prerequisites I used R-3.6.3 and RStudio Preview 1.4. I also plan to use Anaconda, Miniconda and GNU Python for the parts where I make use of Python code.

How deep should I go in learning data science, machine learning and computer science?

18 November 2020

data science, machine learning

How deep should I go in learning data science, machine learning and computer science? First of all, of course, we cannot lose focus of [petroleum] engineering. That’s what makes us “the Domain Experts” or SMEs. But this new industrial revolution based on data we are living in, requires a new set of lenses to understand, and discover things that were not so evident few years ago. I am not saying you should turn in a professional programmer, and neither to be an amateur; it is about learning the bare minimum to be able to state a problem, describe the workflow, discuss with DS and ML experts, and work out a prototype before scaling it.

rTorch, a minimal book

26 October 2020

machine learning

A Minimal Book for rTorch Why do we want a package of something that is already working well, such as PyTorch? There are several reasons, but the main one is to bring another machine learning framework to R. Probably, it is just me but I feel PyTorch very comfortable to work with. Feels pretty much like everything else in Python. Very pythonic. I have tried other frameworks in R. The closest that matches a natural language like PyTorch, is MXnet.

rTorch - a PyTorch wrapper in R

16 October 2020

machine learning

rTorch The goal of rTorch is providing an R wrapper to PyTorch. rTorch provides all the functionality of PyTorch plus all the features that R provides. We have borrowed some ideas and code used in R tensorflow to implement rTorch. Besides the module torch, which directly provides PyTorch methods, classes and functions, the package also provides the modules numpy as a method called np, and torchvision, as well. The dollar sign $ after the module will provide you access to all their sub-objects.

The introductory tutorial to my package rTorch has just been released

14 October 2020

machine learning

The introductory tutorial to #rTorch has just been released. You can read it here: https://lnkd.in/g8Fvre4, or access the whole code to the book written in #Rmarkdown over here: https://lnkd.in/g6iFT_u. This time I am including code in R and Python. The code chunks for Python are greenish, and the #rstats code in blueish. You will also find some code run using the #Nix terminal; that code has gray background. There are several examples on basic tensor operations and couple of neural networks to get you started.

My R package rTorch 0.4.2 accepted by CRAN

12 October 2020

machine learning

My R package #rTorch 0.4.2 accepted by #rstats CRAN. CRAN link: https://lnkd.in/fw-R-sE rTorch is a series of bindings to #Python machine learning platform #PyTorch, so it makes easier to build neural networks, GANs, RNNs, within the R environment. The link to the repository is https://lnkd.in/fuvNPcv. There is also an introductory tutorial (and package manual) at https://lnkd.in/gFQnDpP. There are two additional manuals I am working on: “The Minimal rTorch Book” written in #Rmarkdown, and “rTorch Advanced Examples”, also using the #rstats #bookdown package.

Debugging rTorch with Docker and Travis

27 August 2020

machine learning

These are some instructions to build and run a Travis machine locally. We attempt to run all the tests usually run at http://travis-ci.org in a Docker container maintaining the same characteristics of a Travis remote machine. Motivation The goal is reducing the time that takes debugging an error in Anaconda environments when installing PyTorch or rTorch. The bugs remain hidden in the local development machine and only pop up in a Travis machine.

Building the book "Statistical Rethinking" by Solomon Kurtz in Docker

13 July 2020

machine learning, statistics

kurtz-rethinking bookdown This book was written by Solomon Kurtz which code lives here. It is a beautiful book on Bayesian regression in R using the brms package. Based on the book “Statistical Rethinking” by Richard McElreath, a Bayesian Course with examples in R and Stan. Bookdown details Book version 1.0.1. See index.Rmd. R-3.6.3 RStudio 1.2.5042 Most packages MRAN dated on 2019-06-12. Other packages dated at later dates for smnoother book building.

Transcript of interview to Stuart Russell by Lex Fridman

19 January 2020

machine learning, artificial intelligence

Stuart Russell: Long-Term Future of Artificial Intelligence | Artificial Intelligence (AI) Podcast | Dec 9, 2018 This is a quick transcript of the interview of Stuart Russell by Lex Fridman. The interview has been published in YouTube as well as a podcast. This is a transcript I wanted to do for a long time. Stuart Russell is an authority on AI and coauthor of one of the best books on Artificial Intelligence.

Transcript of interview of Ian Goodfellow by Lex Fridman

24 October 2019

artificial intelligence, machine learning

Ian Goodfellow: Generative Adversarial Networks (GANs) | MIT Artificial Intelligence (AI) Podcast | April 18, 2019 This is a quick transcript of the interview of Ian Goodfellow by Lex Fridman. The interview has been published in YouTube as well as a podcast. Goodfellow points at so many areas in the future where ML and AI could bring astonishing discoveries. Enjoy! Alfonso R. Reyes. The Woodlands, Texas. Interview to Ian Goodfellow The following is a conversation with Ian Goodfellow.

All Blog posts by date