guy4261

#!/usr/bin/env python
# coding: utf-8
import os
import webbrowser
import zipfile
from datetime import datetime
from glob import glob
from io import BytesIO

import pandas as pd

guy4261

# To create a hebrew letter with niqqud:
# hebrew letter [+ optional shin_dot if letter is shin] [+ optional dagesh] [+ optional niqqud]
# example: print("ש" + chr(shin_dot_right_shin) + chr(dagesh) + chr(kmz_katan)) => שָּׁ

# letter should be first, order of the rest does not matter
# print("ש" + chr(kmz_katan) + chr(shin_dot_left_sinn) + chr(dagesh)) => שָּׂ

# This is how you get things like the reverse of noël being l̈eon instead of lëon,
# as discussed in Edaqa Mortoray's two seminal blog posts:
# https://mortoray.com/we-dont-need-a-string-type/

guy4261

# Download an install miniconda
wget -nv https://repo.continuum.io/miniconda/Miniconda-latest-Linux-x86_64.sh
chmod u+x Miniconda-latest-Linux-x86_64.sh
./Miniconda-latest-Linux-x86_64.sh -b -f


# -b means batch install using the defaults: that means .bashrc does not change so we do it outselves.
cp .bashrc .bashrc-miniconda.bak
echo "
# added by Miniconda 3.10.1 installer

guy4261

###########################
# Installing Python 2.7.9 #
###########################

# https://github.com/h2oai/h2o-2/wiki/Installing-python-2.7-on-centos-6.3.-Follow-this-sequence-exactly-for-centos-machine-only
yum -y groupinstall "Development tools"

yum -y install zlib-devel
yum -y install bzip2-devel
yum -y install openssl-devel

guy4261

train = torch.load('mnist.t7/train_32x32.t7', 'ascii')
train.data = train.data:type(torch.getdefaulttensortype())

function pixels(dataset)
    local dimensions = (#dataset.data):totable()    
    local data = dataset.data
    local d1 = 1
    local d2 = 1
    local d3 = 1
    local d4 = 1

guy4261

require('torch')
-- Find the average per image
train = torch.load('mnist.t7/train_32x32.t7', 'ascii')
train.data = train.data:type(torch.getdefaulttensortype())
BOUND = 60000 -- Perform on the full data
-- BOUND = 1 -- You can test on a single image as a beginning
for index = 1, BOUND do
    cells = train.data[index][1]
    avg = 0
    

guy4261

On mouseover, mark the node you're hovering over, and color its neighbours according to the type of the relationship. This style is only an example and anything goes - from adding icons, respecting asymmetric links (i.e. different markup of employee or manager in an unmutual hierarchy link) and the list goes on.

Not relying on markers etc gives a lot of visual relief.

Based on Mike Bostock's molecule example from: http://bl.ocks.org/mbostock/3037015

guy4261

<?php
header('Content-type: text/plain');
header('Content-Disposition: attachment; filename="default-filename.txt"');

$newline = "\r\n";
$delimiter = "\t";

$db_host = 'localhost';
$db_user = "web";
$db_pwd = '';