chrishwiggins

repo structure summary

( please email chris.wiggins@gmail.com with questions, comments, or suggestions )

• aux: auxiliary files, facts and useful things meant for human eyes
• dat: data, not for human eyes
• doc: things you edit for humans to read
• eml: emails sent and received
• fig: figures
• git: git or GitHub repos

chrishwiggins

% thoughts on deep, 2016-09-01
% @chrishwiggins

(apologies this ended up being long; most of the ideas are in 1 graphic image,
so feel free to just click on the link 
( https://sketch.io/render/sk-e40f367014c9440fef81de46271b4395.jpeg )
and you'll get the main ideas in about 1-2 seconds, and can save
the text for sometime when you're stuck in an elevator)

I was thinking about how deep learning as a capability relates to a company's challenges.

chrishwiggins

( advice from @johnmyleswhite; ministorm starts with https://twitter.com/johnmyleswhite/status/766657182582513666 )


I keep planning on writing a post about contributing to OSS effectively, but I'm starting to doubt I ever will. So here's some tweets.
(1) Credibility is everything. People are busy, so they heavily use personal trust to inform their decisions. You must become credible.
(2) Credibility starts with technical competence, rather than shared aesthetics. Do *not* start contributing by engaging in design debates.
(3) Instead, work on simple projects where success or failure will be clear. Write unit tests. Write docs. Improve a function's performance.
(4) That kind of work will allow you to demonstrate skill, which will make you credible. After, people will want to hear your design goals.
(5) Don't engage in debates that are long-standing without being familiar with the entire past history of the debate.
(6) Especially don't focus on the shallow part of a debate where you can participate without kno

chrishwiggins

Finally posting my 389-page tutorial 
"data science @ the new york times"
from last week's Machine Learning Summer School
in Arequipa, Peru ( http://mlss.cc/ )
here: http://www.slideshare.net/chrishwiggins/machine-learning-summer-school-2016

Topics include:

descriptive/predictive/prescriptive modeling
(unsupervised/supervised/reinforcement learning),

chrishwiggins

how to write a paper (one possible answer)

Originally appeared on nature NYC blog: http://blogs.nature.com/nyc/2011/08/10/how-to-write-a-paper-one-possible-answer

10 Aug 2011 | 19:11 EDT | Posted by chris wiggins | Category: Uncategorized

how to write a paper

a student recently asked me how to write a paper. here’s an algorithm i’d suggest, with plenty of room for an individual to deviate.

chrishwiggins

2007: next-generation sequencing.

2008:

With its tremendous potential for understanding cellular biology now poised to become a reality, super-resolution fluorescence microscopy is our choice for Method of the Year.

2009:

Nature Methods' Method of the Year 2009 goes to induced pluripotency for its potential for biological discovery.

2010:

With the capacity to control cellular behaviors using light and genetically encoded light-sensitive proteins, optogenetics has opened new doors for experimentation across biological fields.

chrishwiggins

data science: references / links (last update 2016)

context

• google search trends: data science

• indeed job trends: data science

chrishwiggins

%%% little code to compute steady state "q" from creation+decay "wp,wm"
% wp=creation
% wm=destruction

function q=qdef(wp,wm);
% 1) most important line p=\Pi (wp_j/wm_{j+1})=S^+_n/S^-/n
p=cumprod(	wp(1:(end-1))	...
			./	...
		wm((1+1):end)	...
);

chrishwiggins

%%% parameters
% forget what you know
clear
% set max N
N=500;
% set "turnover n", aka n0
n0=10;
% plotting parameters:
zmax=.2;
xmax=60;

chrishwiggins

import sys
import random
u_Health = 10
t_Health = 10
while u_Health*t_Health > 0:
    they_attack=random.randint(0,1)
    if they_attack == 0:
        print ("they attack!")
        u_Health=u_Health-1
        #print (u_Health,t_Health)