timvieira

import numpy as np
import pylab as pl
from numpy import exp, cos, sin, sqrt
from arsenal.math import compare

def run_tests():
    tests = """
x**2
2*x

timvieira

# -*- coding: utf-8 -*-
"""
Reversing a singly-linked sequence defined by a function application in
sublinear space.

  s[t] = f(s[t-1]) where s[0] is given as input.

Code associated with blog post
"Reversing a sequence with sublinear space"
http://timvieira.github.io/blog/post/2016/10/01/reversing-a-sequence-with-sublinear-space/

timvieira

function yellow { echo -e "\e[33m$@\e[0m"; }

# Wrapper around make, which covers building different project types, when an
# actual Makefile isn't present.
function make {
    if [[ -e Makefile ]]; then
        yellow "[make] found Makefile"
        /usr/bin/make $@
    else
        yellow "[make] No Makefile found"

timvieira

import numpy as np
from numpy.random import uniform


def update(S, k, v):
    "Update value position `k` in time O(log n)."
    d = S.shape[0]
    i = d//2 + k
    S[i] = v
    while i > 0:

timvieira

timvieira

"""
Cartoon version of Jiawei's optimization problem.

Created [2017-02-17 Fri]

"""
import numpy as np
from scipy.optimize import fmin_bfgs

import autograd

timvieira

"""
Memory-efficient backpropagation in an RNN.

Accompanies blog post:
http://timvieira.github.io/blog/post/2016/10/01/reversing-a-sequence-with-sublinear-space/
"""

import numpy as np
from arsenal.math.checkgrad import fdcheck

timvieira

"""
Memory-efficient backpropagation in an RNN.

Accompanies blog post:
http://timvieira.github.io/blog/post/2016/10/01/reversing-a-sequence-with-sublinear-space/
"""

import numpy as np
from arsenal.math.checkgrad import fdcheck

timvieira

# Efficient passive aggressive updates for multi-class classification
#
# Original article:
# "Column squishing for multiclass updates"
# https://nlpers.blogspot.com/2017/08/column-squishing-for-multiclass-updates.html


from __future__ import division
import numpy as np
import scipy.optimize

timvieira

"""
Simple example of manually performing "automatic" differentiation
"""
import numpy as np
from numpy import exp, sin, cos


def f(x, with_grad=False):
    # Need to cache intermediates from forward pass (might not use all of them).
    a = exp(x)