jcboyd

# $ docker run \
#   --device /dev/nvidia0:/dev/nvidia0 \
#   --device /dev/nvidiactl:/dev/nvidiactl \
#   --device /dev/nvidia-uvm:/dev/nvidia-uvm jcboyd/ubuntu-cuda-caffe

# Download base image
FROM ubuntu:16.04

# Set environment variables
ENV PATH "/usr/local/cuda-8.0/bin:$PATH"

jcboyd

jcboyd

import os
import cStringIO
import requests
import wave


# acquire credentials at https://www.ibm.com/watson/services/text-to-speech/
user_name = ''
password = ''

jcboyd

from __future__ import print_function
from __future__ import division

import io
import random
import numpy as np
from PIL import Image

from skimage.transform import resize
import quantumrandom  # https://qrng.anu.edu.au/

jcboyd

jcboyd

jcboyd

jcboyd

\documentclass[tikz]{standalone}
\usepackage{tikz}

\begin{document}

\def\primes{2,3,5,7,11,13,17,19,23}
\def\revprimes{23,19,17,13,11,7,5,3,2}

\def\palette{{
 "0.00 0.00 1.00",

jcboyd

"""Two tricks enable an efficient solution:
2. We can solve the problem for the first i numbers in the list, for
   i = 1, ..., N
1. The range of possible sums, R is bounded by an efficiently computable
   min_sum and max_sum.
The lookup table is thus of dimension N x R. For each subset i = 1, ..., N,
we determine the range of values attainable with that subset, which are
simply all the targets of subset i - 1, plus nums[i] (included), and plus 0
(not included). Thus, we perform only R operation per N subsets.
"""

jcboyd

import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns

from matplotlib.patches import Circle, Rectangle, FancyArrowPatch
import matplotlib.lines as mlines

from scipy.ndimage import gaussian_filter