miku

# How to create an RPM repository

# This is for Redhat 64 bit versions of Linux. You can create your own RPM repository # to host your custom RPM packages.
#
# See "How to create an RPM from source with spec file" for more information.
# https://gist.github.com/1376973

# Step: 1
# Install createrepo

miku

#!/bin/sh

# This is for Redhat 64 bit versions of Linux with `createrepo` installed. If you
# do not have createrepo, you can install it with:
# yum install -y createrepo

# Change DESTDIR path to RPMS directory of your repo
DESTDIR="/var/www/repo/rhel/6"

for ARCH in x86_64

miku

% From: https://docs.google.com/file/d/0B6hnQC2sJnn-R2RaZ2gybUNaQjg/edit?pli=1
% \usepackage[usenames,dvipsnames]{pstricks}
% \usepackage{epsfig}
% \usepackage{pst-grad} % For gradients
% \usepackage{pst-plot} % For axes
% User Packages:
% \usepackage{arial}
\psscalebox{1.0 1.0} % Change this value to rescale the drawing.
{
\begin{pspicture}(0,-9.555931)(28.244461,9.555931)

miku

<!-- 
Visual comparison: http://imgur.com/a/Fdwxp

* Basic display.
* No multiple tags, multiple subfield values.
* No "Show more".
* 25 LOC.
-->
<script type="text/javascript">
    $(document).ready(function() {

miku

# just the master
curl localhost:9200/_cat/master

# all nodes
curl localhost:9200/_cat/nodes

# all indices
curl localhost:9200/_cat/indices

# (bsz) shards ok (STARTED)?

miku

#
# It's not really elegant (I'm happy to hear about alternatives),
# but has done what it should so far. 
#
# lftp takes care of the mirroring and this task is completed if a single file containing
# the paths to the mirrored file has been successfully written - so it's a bit indirect,
# but at least follows the one-file-per-task rule.
#
# It would be ok to run this task over and over, since after the first transfer the directories
# should be in sync.

miku

Story

A conditional probability is a probability based on some background information - that make up a condition.

The cookie problem

miku

* Resilience instead of strength
* Pull, instead of push
* Take risks, instead of safety
* System, instead of objects
* God compasses, not maps
* Work on practice, instead of theory
* Disobedience, instead of compliance
* Crowd, instead of the experts
* Learning, instead of education

miku

#!/usr/bin/env python

def max_subarray(A):
    max_ending_here = max_so_far = 0
    for x in A:
        max_ending_here = max(0, max_ending_here + x)
        max_so_far = max(max_so_far, max_ending_here)
    return max_so_far

if __name__ == '__main__':

miku

Notes

Basic problems: NN and ANN.

Locality sensitive hashing is a family of algorithms for NN.

Given a collection of n points, build a data structure which, given any query point, reports the data point, that is closest to the query.