Today I Learned

2019-03-23

Markov models

A Markov model is a stochastic model used to model randomly changing systems where it is assumed that future states depend only on the current state not on the events that occurred before it (that is, it assumes the Markov property). Generally, this assumption enables reasoning and computation with the model that would otherwise be intractable.

• From “What is a Markov Model” to “Here is how Markov Models Work”
• Markov model

2019-03-14

How to split a CSV file using split

split is a Unix command used to split a file into pieces. It can be used like this:

Copycopy code to clipboard
split -l 100000 large.csv segment

The command above splits a file with a limit of 100k lines each. If the file is 300k then it will create 3 files.

You can also use a maximum number of bytes by using the -b flag:

Copycopy code to clipboard
split -b 40k large.csv segment

2019-01-26

• How Support Vector Machines work:
  1. Support Vector Machine — Introduction to Machine Learning Algorithms
  2. Chapter 2 : SVM (Support Vector Machine) — Theory

2019-01-14

Clean Git Repository with Tracked and Untracked file

Copycopy code to clipboard
git checkout master
git clean -dxf
git reset --hard

2019-01-10

Build to Learn by Jeff Patton

• Building to learn means doing things as cheap experiments. Once you’ve learned that users actually want the feature, you can then spend full engineering rigor toward building to earn. You will likely end up throwing away your “learn” code.
• Building software is one of the most expensive ways to learn.
• Shipping small things is always easier.
• Code is a liability. Knowledge is an asset.
• Directed Discovery is a form of “building to learn”. We don’t have to build software to learn. We can use prototypes or wireframes.

2019-01-09

• Basics of Natural Language Processing.
• spaCy and textacy NLP libraries.
• Sourcegraph: Code search and intelligence