tansey

import tensorflow as tf

def unpack_cholesky(q, ndims):
    # Build the lower-triangular Cholesky from the flat buffer (assumes q shape is [batchsize, cholsize])
    chol_diag = tf.nn.softplus(q[:,:ndims])
    chol_offdiag = q[:,ndims:]
    chol_rows = []
    chol_start = 0
    chol_end = 1
    for i in xrange(ndims):

tansey

#include <set>
#include <iostream>

class A
{
    std::set<int> s;
public:
    A() { s.insert(0); s.insert(1); }

    std::set<int> getSet() { return s; }

tansey

'''
Implementation of the ADMM convergence rate SDP from Nishihara et al.,
ICML 2015, equation 11.

Code by Wesley Tansey and Sanmi Koyejo
7/31/2015
'''
import cvxpy as cvx
import numpy as np

tansey

function format(s, args...)
    # Python-style string formatting with floating point support
    # Note that this is 1-based to be more Julian
    result = deepcopy(s)
    for (i, x) in enumerate(args)
        q = Regex("{$i(:\.([0-9])+f)?}")
        next = result
        for m in eachmatch(q, result)
            val = x
            if m.captures[2] != nothing

tansey

import numpy as np
import numpy.ma as ma

# Create a 3x3 array in regular numpy
a = np.arange(9).reshape((3,3))

# Get the middle row
b = a[1]

# Change the middle value in the middle row

tansey

def plot_with_bands(graph_title, means, bands, series, xvals=None, xlabel=None, ylabel=None, subtitle=None, filename='results.pdf'):
    colors = ['blue','red','green', 'black', 'yellow', 'orange', 'purple', 'brown'] # max 8 lines
    print 'means: {0} bands: {1}'.format(means.shape, bands.shape)
    assert(means.shape == bands.shape)
    assert(xvals is None or xvals.shape[0] == means.shape[1])
    assert(means.shape[1] <= len(colors))
    if xvals is None:
        xvals = np.arange(means.shape[0])
    ax = plt.subplot(111)
    plt.ticklabel_format(axis='y', style='plain', useOffset=False)

tansey

import random
import math
import matplotlib.pyplot as plt
import numpy as np

def acf(x, length=35):
  return np.array([1] + [np.corrcoef(x[:-i], x[i:])[0,1] for i in range(1,length)])

# The number of samples to observe
n = 100

tansey

class A(object):
  def __init__(self, myvar = {}):
    self.myvar = myvar
    
a1 = A()
a1.myvar['foo'] = 1
 
a2 = A()
a2.myvar['bar'] = 2
 

tansey

class A(object):
  def __init__(self, myvar = {}):
    self.myvar = myvar
    
a1 = A()
a1.myvar['foo'] = 1

a2 = A()
a2.myvar['bar'] = 2

tansey

double[] inputs = ...
double[] desiredOutputs = ...

// Get the neural network
var network = ((FastCyclicNetwork)blackbox;

// Perform backpropagation through time
network.Train(inputs, desiredOutputs);