chryss

import argparse
import collections

# Let's define a global dispatch table for the 8 different possible neighbors
# clockwise starting top-left: tl, t, tr, r, br, b, bl, l.
# The target index only depends on the number of rows and the start index jj
# As we won't know beforehand which functions to use for a given start index,
# we make them all dependent on both jj and n

neighbordispatch = {

chryss

import argparse

def dosomething(someinput):
    pass
    
def main():
    parser = argparse.ArgumentParser()
    parser.add_argument("infilepath")
    args = parser.parse_args()

chryss

#!/usr/bin/env python
# encoding: utf-8
"""
evaluatehfdi_debug.py

Debugging scikit-learn Gaussian Mixture Model parameter choice.

Created by Chris Waigl on 2014-05-13.
Copyright (c) 2014 Christine F. Waigl. MIT License.
"""

chryss

#!/usr/bin/env bash
#
# clip_usgs_to_shape.sh
# Clip individual band files from USGS-distributed remote-senisng images
# to extent of polygon provided in a shapefile, using the GDAL command line
# tools.
#
# Chris Waigl, 2014-04-15
#
# Input:

chryss

{
 "metadata": {
  "name": ""
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {

chryss

chryss

import json
import numpy as np
from osgeo import gdal, gdal_array

typemap = {}
typenames = {}
for name in dir(np):
    obj = getattr(np, name)
    if hasattr(obj, 'dtype'):
        try:

chryss

require(RCurl)
require(reshape2)
require(ggplot2)

puppies <- getURL("https://docs.google.com/spreadsheet/pub?key=0AtKrmILNPprEdEY4MFA5SWt5THRyYlB4cnhUNkxuOVE&single=true&gid=3&output=csv")
puppydat <- read.csv(textConnection(puppies), header=TRUE)
puppyflat <- melt(puppydat)

p <- ggplot(puppyflat, aes(X, value, color=factor(variable), group=variable))

chryss

chryss

{
 "metadata": {
  "name": ""
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {