kcarnold

import numpy as np
from sklearn.datasets import fetch_20newsgroups
from sklearn.feature_extraction.text import CountVectorizer

def get_vectors(vocab_size=5000):
    newsgroups_train = fetch_20newsgroups(subset='train')
    vectorizer = CountVectorizer(max_df=.9, max_features=vocab_size)
    vecs = vectorizer.fit_transform(newsgroups_train.data)
    vocabulary = vectorizer.vocabulary
    terms = np.array(vocabulary.keys())

kcarnold

Google Apps Script is clunky. You have to make a spreadsheet in Google Docs, then Tools->Script Editor. Then you can paste this code in a new project. Try running it once to make sure it works, then under Triggers select moveRandomSnoozes, time-based, and whatever frequency you want. (I had it going daily, but just changed the probabilities and set it to hourly; we'll see how I like that.)

This was partly based on the Gmail Snooze script that was making the rounds a few months ago.

kcarnold

# Dirichlet process Gaussian mixture model

import numpy as np
from scipy.special import gammaln
from scipy.linalg import cholesky
from sliceSample import sliceSample

def multinomialDraw(dist):
    """Returns a single draw from the given multinomial distribution."""
    return np.random.multinomial(1, dist).argmax()

kcarnold

% All based on Rasmussen and Williams.

N = 1000;
lo=0; hi=5;
x = linspace(lo, hi, N);
%xn = [2, 2.5, 3]';
%yn = [-1.9, -2, -1.9]';
xn = ([-4, -3, -1, 0, 2]'+5)/2;
yn =  [-2,  0,  1, 2, -1]';

kcarnold

var Event = (function() {
  function Event(name) {
    this.name = name;
    this.listeners = [];
  }
  Event.prototype.when = function(context, callback) {
    if (arguments.length === 1) {
      callback = context;
      context = null;
    }

kcarnold

function logBrowserStats() {
  function getAsObject(objectLike) {
    var obj = {}, key;
    for (key in objectLike) {
      var val = objectLike[key];
      if (typeof val === 'number' || typeof val === 'string') {
        obj[key] = val;
      }
    }
    return obj;

kcarnold

import sys
tr = [
    (u'\u2018', "`"),
    (u'\u2019',"'"),
    (u'\u201c', "``"),
    (u'\u201d', "''")
    ]
s = sys.stdin.read().decode('utf8')
for a, b in tr:
    s = s.replace(a, b)

kcarnold

import lxml.html
import sys

doc = lxml.html.fromstring(sys.stdin.read())

for elt in doc.cssselect('a, div, style, title'):
    elt.getparent().remove(elt)

s = u'\n'.join((elt.text_content() for elt in doc.cssselect('p, h1, h2, h3, h4, h5, h6'))) # add other nodes if I forgot any

kcarnold

# ~/.bash_profile: executed by bash(1) for login shells.

source .bashrc

kcarnold

# don't put duplicate lines in the history. See bash(1) for more options
HISTCONTROL=ignoredups:ignorespace
HISTIGNORE=ls:fg
# append to the history file, don't overwrite it
shopt -s histappend

# for setting history length, see HISTSIZE and HISTFILESIZE in bash(1)
HISTSIZE=10000
HISTFILESIZE=20000