ramhiser

cut_mixture <- function(x, range = NULL, tol = 1e-6, maxit = 100) {
  n <- length(x)
  
  if (is.null(range)) {
    range <- c(min(x), max(x))
  }
  
  # Initialization
  # Initial value of pi1 are the proportion of x within 2 standard deviations of mu
  huber_out <- huber(x)

ramhiser

set.seed(42)

n <- 50

x1 <- rnorm(n, mean = 5)
x2 <- rnorm(n, mean = 0)
x3 <- rnorm(n, mean = 20)

plot(density(x1), xlim = c(-5, 25))
lines(density(x2), col = "red")

ramhiser

normalmix_loglike <- function(x, y) {
  y <- as.factor(y)
  x1 <- x[y == levels(y)[1]]
  x2 <- x[y == levels(y)[2]]
  
  n1 <- length(x1)
  n2 <- length(x2)
  n <- n1 + n2
  
  w1 <- n1 / n

ramhiser

# The following script scrapes ESPN's MLB Standings Grid and writes the
# standings for each American League (AL) team to a CSV file, which has the following
# format:
# Team, Opponent, Wins, Losses

from bs4 import BeautifulSoup
import urllib2
import re
import csv

ramhiser

Interactive Visualization of CyTOF Data
========================================================

```{r setup, echo=FALSE}
library(rgl)
knit_hooks$set(webgl = hook_webgl)
opts_knit$set(upload.fun = imgur_upload, base.url = NULL) # upload all images to imgur.com
```

This report produces a 3D visualization of the CD8+ T-cell subsets from [Newell et al. (2012)](http://www.ncbi.nlm.nih.gov/pubmed/22265676) using principal components analysis (PCA).

ramhiser

import nltk
from nltk.collocations import *
from nltk.book import *
import re

bigram_measures = nltk.collocations.BigramAssocMeasures()

# Monty Python and the Holy Grail

# Reduces tokens to words. Ignores ALL CAPS words, which are the speaker in the movie.

ramhiser

# For info about census regions, see:
# http://en.wikipedia.org/wiki/List_of_regions_of_the_United_States#Census_Bureau-designated_regions_and_divisions

# Region - Northeast

# Division - New England
new_england <- data.frame(
  region = "Northeast",
  division = "New England",
  state = c("ME", "NH", "VT", "MA", "RI", "CT")

ramhiser

import numpy as np
import pandas as pd
from itertools import chain, izip, repeat

np.random.seed(42)
num_rows = 10
num_features = 5
num_feature_values = 3

# Builds tuples of features with many values per feature

ramhiser

library(dplyr)
group_size <- 20
foo <- iris[1:119, ]
filter(group_by(foo, Species), n() >= group_size)

ramhiser

# FCC's Census Block Conversions API
# http://www.fcc.gov/developers/census-block-conversions-api
latlong2fips <- function(latitude, longitude) {
  url <- "http://data.fcc.gov/api/block/find?format=json&latitude=%f&longitude=%f"
  url <- sprintf(url, latitude, longitude)
  json <- RCurl::getURL(url)
  json <- RJSONIO::fromJSON(json)
  as.character(json$County['FIPS'])
}