Motivation: I have a multidimensional integral, which for completeness I have reproduced below. It comes from the computation of the second virial coefficient when there is significant anisotropy:
Here W is a function of all the variables. It is a known function, one which I can define a python function for.
Programming Question: How do I get scipy to integrate this expression? I was thinking of chaining two triple quads (scipy.integrate.tplquad) together, but I'm worried about performance and accuracy. Is there a higher dimensional integrator in scipy, one that can handle an arbitrary number of nested integrals? If not, what is the best way to do this?
With a higher-dimensional integral like this, monte carlo methods are often a useful technique - they converge on the answer as the inverse square root of the number of function evaluations, which is better for higher dimension then you'll generally get out of even fairly sophisticated adaptive methods (unless you know something very specific about your integrand - symmetries that can be exploited, etc.)
The mcint package performs a monte carlo integration: running with a non-trivial W that is nonetheless integrable so we know the answer we get (note that I've truncated r to be from [0,1); you'll have to do some sort of log transform or something to get that semi-unbounded domain into something tractable for most numerical integrators):
import mcint
import random
import math
def w(r, theta, phi, alpha, beta, gamma):
return(-math.log(theta * beta))
def integrand(x):
r = x[0]
theta = x[1]
alpha = x[2]
beta = x[3]
gamma = x[4]
phi = x[5]
k = 1.
T = 1.
ww = w(r, theta, phi, alpha, beta, gamma)
return (math.exp(-ww/(k*T)) - 1.)*r*r*math.sin(beta)*math.sin(theta)
def sampler():
while True:
r = random.uniform(0.,1.)
theta = random.uniform(0.,2.*math.pi)
alpha = random.uniform(0.,2.*math.pi)
beta = random.uniform(0.,2.*math.pi)
gamma = random.uniform(0.,2.*math.pi)
phi = random.uniform(0.,math.pi)
yield (r, theta, alpha, beta, gamma, phi)
domainsize = math.pow(2*math.pi,4)*math.pi*1
expected = 16*math.pow(math.pi,5)/3.
for nmc in [1000, 10000, 100000, 1000000, 10000000, 100000000]:
random.seed(1)
result, error = mcint.integrate(integrand, sampler(), measure=domainsize, n=nmc)
diff = abs(result - expected)
print "Using n = ", nmc
print "Result = ", result, "estimated error = ", error
print "Known result = ", expected, " error = ", diff, " = ", 100.*diff/expected, "%"
print " "
Running gives
Using n = 1000
Result = 1654.19633236 estimated error = 399.360391622
Known result = 1632.10498552 error = 22.0913468345 = 1.35354937522 %
Using n = 10000
Result = 1634.88583778 estimated error = 128.824988953
Known result = 1632.10498552 error = 2.78085225405 = 0.170384397984 %
Using n = 100000
Result = 1646.72936 estimated error = 41.3384733174
Known result = 1632.10498552 error = 14.6243744747 = 0.8960437352 %
Using n = 1000000
Result = 1640.67189792 estimated error = 13.0282663003
Known result = 1632.10498552 error = 8.56691239895 = 0.524899591322 %
Using n = 10000000
Result = 1635.52135088 estimated error = 4.12131562436
Known result = 1632.10498552 error = 3.41636536248 = 0.209322647304 %
Using n = 100000000
Result = 1631.5982799 estimated error = 1.30214644297
Known result = 1632.10498552 error = 0.506705620147 = 0.0310461413109 %
You could greatly speed this up by vectorizing the random number generation, etc.
Of course, you can chain the triple integrals as you suggest:
import numpy
import scipy.integrate
import math
def w(r, theta, phi, alpha, beta, gamma):
return(-math.log(theta * beta))
def integrand(phi, alpha, gamma, r, theta, beta):
ww = w(r, theta, phi, alpha, beta, gamma)
k = 1.
T = 1.
return (math.exp(-ww/(k*T)) - 1.)*r*r*math.sin(beta)*math.sin(theta)
# limits of integration
def zero(x, y=0):
return 0.
def one(x, y=0):
return 1.
def pi(x, y=0):
return math.pi
def twopi(x, y=0):
return 2.*math.pi
# integrate over phi [0, Pi), alpha [0, 2 Pi), gamma [0, 2 Pi)
def secondIntegrals(r, theta, beta):
res, err = scipy.integrate.tplquad(integrand, 0., 2.*math.pi, zero, twopi, zero, pi, args=(r, theta, beta))
return res
# integrate over r [0, 1), beta [0, 2 Pi), theta [0, 2 Pi)
def integral():
return scipy.integrate.tplquad(secondIntegrals, 0., 2.*math.pi, zero, twopi, zero, one)
expected = 16*math.pow(math.pi,5)/3.
result, err = integral()
diff = abs(result - expected)
print "Result = ", result, " estimated error = ", err
print "Known result = ", expected, " error = ", diff, " = ", 100.*diff/expected, "%"
which is slow but gives very good results for this simple case. Which is better is going to come down to how complicated your W is and what your accuracy requirements are. Simple (fast to evaluate) W with high accuracy will push you to this sort of method; complicated (slow to evaluate) W with moderate accuracy requirements will push you towards MC techniques.
Result = 1632.10498552 estimated error = 3.59054059995e-11
Known result = 1632.10498552 error = 4.54747350886e-13 = 2.7862628625e-14 %
Jonathan Dursi has made a very good answer. I will just add on to his answer.
Now scipy.integrate has a function named nquad that one can perform a multi-dimensional integral without hassle. See this link for more information. Below we compute the integral using nquad with Jonathan's example:
from scipy import integrate
import math
import numpy as np
def w(r, theta, phi, alpha, beta, gamma):
return(-math.log(theta * beta))
def integrand(r, theta, phi, alpha, beta, gamma):
ww = w(r, theta, phi, alpha, beta, gamma)
k = 1.
T = 1.
return (math.exp(-ww/(k*T)) - 1.)*r*r*math.sin(beta)*math.sin(theta)
result, error = integrate.nquad(integrand, [[0, 1], # r
[0, 2 * math.pi], # theta
[0, math.pi], # phi
[0, 2 * math.pi], # alpha
[0, 2 * math.pi], # beta
[0, 2 * math.pi]]) # gamma
expected = 16*math.pow(math.pi,5)/3
diff = abs(result - expected)
The result is more accurate than the chained tplquad:
>>> print(diff)
0.0
I'll just make a couple of general comments about how to accurately do this sort of integral, but this advice is not specific to scipy (too long for a comment, even though it is not an answer).
I don't know your use case, i.e. whether you are satisfied with a `good' answer with a few digits of accuracy which could be obtained straightforwardly using Monte Carlo as outlined in Jonathan Dursi's answer, or whether you really want to push the numerical accuracy as far as possible.
I've performed analytic, Monte Carlo and quadrature calculations of virial coefficients myself. If you want to do the integrals accurately, then there are a few things you should do:
Attempt to perform as many of the integrals exactly as possible; it may well be that integration in some of your coordinates is quite simple.
Consider transforming your variables of integration so that the integrand is as smooth as possible. (This helps for both Monte Carlo and quadrature).
For Monte Carlo, use importance sampling for best convergence.
For quadrature, with 7 integrals it may just be possible to get really fast convergence using tanh-sinh quadrature. If you can get it down to 5 integrals then you should be able to get 10's of digits of precision for your integral. I highly recommend mathtool / ARPREC for this purpose, available from David Bailey's homepage: http://www.davidhbailey.com/
First to say that I am not that good in math so please be kind. Anyway, here is my try:
Note that in your question there are 6 variables but 7 integrals!?
In Python using Sympy:
>>> r,theta,phi,alpha,beta,gamma,W,k,T = symbols('r theta phi alpha beta gamma W k T')
>>> W = r+theta+phi+alpha+beta+gamma
>>> Integral((exp(-W/(k*T))-1)*r**2*sin(beta)*sin(theta),(r,(0,2*pi)),(theta,(0,pi)),(phi,(0,2*pi)),(alpha,(0,2*pi)),(beta,(0,pi)),(gamma,(0,pi)))
>>> integrate((exp(-W)-1)*r**2*sin(beta)*sin(theta),(r,(0,2*pi)),(theta,(0,pi)),(phi,(0,2*pi)),(alpha,(0,2*pi)),(beta,(0,pi)),(gamma,(0,pi)))
and here is the result: [LateX code]
\begin{equation*}- \frac{128}{3} \pi^{6} - \frac{\pi^{2}}{e^{2 \pi}} - \frac{\pi}{e^{2 \pi}} - \frac{2}{e^{2 \pi}} - \frac{\pi^{2}}{e^{3 \pi}} - \frac{\pi}{e^{3 \pi}} - \frac{2}{e^{3 \pi}} - 3 \frac{\pi^{2}}{e^{6 \pi}} - 3 \frac{\pi}{e^{6 \pi}} - \frac{2}{e^{6 \pi}} - 3 \frac{\pi^{2}}{e^{7 \pi}} - 3 \frac{\pi}{e^{7 \pi}} - \frac{2}{e^{7 \pi}} + \frac{1}{2 e^{9 \pi}} + \frac{\pi}{e^{9 \pi}} + \frac{\pi^{2}}{e^{9 \pi}} + \frac{1}{2 e^{8 \pi}} + \frac{\pi}{e^{8 \pi}} + \frac{\pi^{2}}{e^{8 \pi}} + \frac{3}{e^{5 \pi}} + 3 \frac{\pi}{e^{5 \pi}} + 3 \frac{\pi^{2}}{e^{5 \pi}} + \frac{3}{e^{4 \pi}} + 3 \frac{\pi}{e^{4 \pi}} + 3 \frac{\pi^{2}}{e^{4 \pi}} + \frac{1}{2 e^{\pi}} + \frac{1}{2}\end{equation*}
You may play a bit more for your question ;)
Related
I have two solutions to this problem actually, they are both applied below to a test case. The thing is that none of them is perfect: first one only take into account the two end points, the other one can't be made "arbitrarily smooth": there is a limit in the amount of smoothness one can achieve (the one I am showing).
I am sure there is a better solution, that kind-of go from the first solution to the other and all the way to no smoothing at all. It may already be implemented somewhere. Maybe solving a minimization problem with an arbitrary number of splines equidistributed?
Thank you very much for your help
Ps: the seed used is a challenging one
import matplotlib.pyplot as plt
from scipy import interpolate
from scipy.signal import savgol_filter
import numpy as np
import random
def scipy_bspline(cv, n=100, degree=3):
""" Calculate n samples on a bspline
cv : Array ov control vertices
n : Number of samples to return
degree: Curve degree
"""
cv = np.asarray(cv)
count = cv.shape[0]
degree = np.clip(degree,1,count-1)
kv = np.clip(np.arange(count+degree+1)-degree,0,count-degree)
# Return samples
max_param = count - (degree * (1-periodic))
spl = interpolate.BSpline(kv, cv, degree)
return spl(np.linspace(0,max_param,n))
def round_up_to_odd(f):
return np.int(np.ceil(f / 2.) * 2 + 1)
def generateRandomSignal(n=1000, seed=None):
"""
Parameters
----------
n : integer, optional
Number of points in the signal. The default is 1000.
Returns
-------
sig : numpy array
"""
np.random.seed(seed)
print("Seed was:", seed)
steps = np.random.choice(a=[-1, 0, 1], size=(n-1))
roughSig = np.concatenate([np.array([0]), steps]).cumsum(0)
sig = savgol_filter(roughSig, round_up_to_odd(n/10), 6)
return sig
# Generate a random signal to illustrate my point
n = 1000
t = np.linspace(0, 10, n)
seed = 45136. # Challenging seed
sig = generateRandomSignal(n=1000, seed=seed)
sigInit = np.copy(sig)
# Add noise to the signal
mean = 0
std = sig.max()/3.0
num_samples = n/5
idxMin = n/2-100
idxMax = idxMin + num_samples
tCut = t[idxMin+1:idxMax]
noise = np.random.normal(mean, std, size=num_samples-1) + 2*std*np.sin(2.0*np.pi*tCut/0.4)
sig[idxMin+1:idxMax] += noise
# Define filtering range enclosing the noisy area of the signal
idxMin -= 20
idxMax += 20
# Extreme filtering solution
# Spline between first and last points, the points in between have no influence
sigTrim = np.delete(sig, np.arange(idxMin,idxMax))
tTrim = np.delete(t, np.arange(idxMin,idxMax))
f = interpolate.interp1d(tTrim, sigTrim, kind='quadratic')
sigSmooth1 = f(t)
# My attempt. Not bad but not perfect because there is a limit in the maximum
# amount of smoothing we can add (degree=len(tSlice) is the maximum)
# If I could do degree=10*len(tSlice) and converging to the first solution
# I would be done!
sigSlice = sig[idxMin:idxMax]
tSlice = t[idxMin:idxMax]
cv = np.stack((tSlice, sigSlice)).T
p = scipy_bspline(cv, n=len(tSlice), degree=len(tSlice))
tSlice = p.T[0]
sigSliceSmooth = p.T[1]
sigSmooth2 = np.copy(sig)
sigSmooth2[idxMin:idxMax] = sigSliceSmooth
# Plot
plt.figure()
plt.plot(t, sig, label="Signal")
plt.plot(t, sigSmooth1, label="Solution 1")
plt.plot(t, sigSmooth2, label="Solution 2")
plt.plot(t[idxMin:idxMax], sigInit[idxMin:idxMax], label="What I'd want (kind of, smoother will be even better actually)")
plt.plot([t[idxMin],t[idxMax]], [sig[idxMin],sig[idxMax]],"o")
plt.legend()
plt.show()
sys.exit()
Yes, a minimization is a good way to approach this smoothing problem.
Least squares problem
Here is a suggestion for a least squares formulation: let s[0], ..., s[N] denote the N+1 samples of the given signal to smooth, and let L and R be the desired slopes to preserve at the left and right endpoints. Find the smoothed signal u[0], ..., u[N] as the minimizer of
min_u (1/2) sum_n (u[n] - s[n])² + (λ/2) sum_n (u[n+1] - 2 u[n] + u[n-1])²
subject to
s[0] = u[0], s[N] = u[N] (value constraints),
L = u[1] - u[0], R = u[N] - u[N-1] (slope constraints),
where in the minimization objective, the sums are over n = 1, ..., N-1 and λ is a positive parameter controlling the smoothing strength. The first term tries to keep the solution close to the original signal, and the second term penalizes u for bending to encourage a smooth solution.
The slope constraints require that
u[1] = L + u[0] = L + s[0] and u[N-1] = u[N] - R = s[N] - R. So we can consider the minimization as over only the interior samples u[2], ..., u[N-2].
Finding the minimizer
The minimizer satisfies the Euler–Lagrange equations
(u[n] - s[n]) / λ + (u[n+2] - 4 u[n+1] + 6 u[n] - 4 u[n-1] + u[n-2]) = 0
for n = 2, ..., N-2.
An easy way to find an approximate solution is by gradient descent: initialize u = np.copy(s), set u[1] = L + s[0] and u[N-1] = s[N] - R, and do 100 iterations or so of
u[2:-2] -= (0.05 / λ) * (u - s)[2:-2] + np.convolve(u, [1, -4, 6, -4, 1])[4:-4]
But with some more work, it is possible to do better than this by solving the E–L equations directly. For each n, move the known quantities to the right-hand side: s[n] and also the endpoints u[0] = s[0], u[1] = L + s[0], u[N-1] = s[N] - R, u[N] = s[N]. The you will have a linear system "A u = b", and matrix A has rows like
0, ..., 0, 1, -4, (6 + 1/λ), -4, 1, 0, ..., 0.
Finally, solve the linear system to find the smoothed signal u. You could use numpy.linalg.solve to do this if N is not too large, or if N is large, try an iterative method like conjugate gradients.
you can apply a simple smoothing method and plot the smooth curves with different smoothness values to see which one works best.
def smoothing(data, smoothness=0.5):
last = data[0]
new_data = [data[0]]
for datum in data[1:]:
new_value = smoothness * last + (1 - smoothness) * datum
new_data.append(new_value)
last = datum
return new_data
You can plot this curve for multiple values of smoothness and pick the curve which suits your needs. You can also apply this method only on a range of values in the actual curve by defining start and end
My objective is to perform an Inverse Laplace Transform on some decay data (NMR T2 decay via CPMG). For that, we were provided with the CONTIN algorithm. This algorithm was adapted to Matlab by Iari-Gabriel Marino, and it works very well. I want to adapt this code into Python. The core of the problem is with scipy.optimize.fmin, which is not minimizing the mean square deviation (MSD) in any way similar to Matlab's fminsearch. The latter results in a good minimization, while the former doesn't.
I have gone through line by line of my adapted code in Python, and the original Matlab. I checked every matrix and every output. I used this to identify that the critical point is in fmin. I also tried scipy.optimize.minimize and other minimization algorithms, but none gave even remotely satisfactory results.
I have made two MWE, for Python and Matlab, to make it reproducible to all. The example data were obtained from the documentation of the matlab function. Apologies if this is long code, but I don't really know how to shorten it without sacrificing readability and clarity. I tried to have the lines match as closely as possible. I am using Python 3.7.3, scipy v1.3.0, numpy 1.16.2, Matlab R2018b, on Windows 8.1. It's a relatively recent Anaconda install (<2 months).
My code:
import numpy as np
from scipy.optimize import fmin
import matplotlib.pyplot as plt
def msd(g, y, A, alpha, R, w, constraints):
""" msd: mean square deviation. This is the function to be minimized by fmin"""
if 'zero_at_extremes' in constraints:
g[0] = 0
g[-1] = 0
if 'g>0' in constraints:
g = np.abs(g)
r = np.diff(g, axis=0, n=2)
yfit = A # g
# Sum of weighted square residuals
VAR = np.sum(w * (y - yfit) ** 2)
# Regularizor
REG = alpha ** 2 * np.sum((r - R # g) ** 2)
# output to be minimized
return VAR + REG
# Objective: match this distribution
g0 = np.array([0, 0, 10.1625, 25.1974, 21.8711, 1.6377, 7.3895, 8.736, 1.4256, 0, 0]).reshape((-1, 1))
s0 = np.logspace(-3, 6, len(g0)).reshape((-1, 1))
t = np.linspace(0.01, 500, 100).reshape((-1, 1))
sM, tM = np.meshgrid(s0, t)
A = np.exp(-tM / sM)
np.random.seed(1)
# Creates data from the initial distribution with some random noise.
data = (A # g0) + 0.07 * np.random.rand(t.size).reshape((-1, 1))
# Parameters and function start
alpha = 1E-2 # regularization parameter
s = np.logspace(-3, 6, 20).reshape((-1, 1)) # x of the ILT
g0 = np.ones(s.size).reshape((-1, 1)) # guess of y of ILT
y = data # noisy data
options = {'maxiter':1e8, 'maxfun':1e8} # for the fmin function
constraints=['g>0', 'zero_at_extremes'] # constraints for the MSD function
R=np.zeros((len(g0) - 2, len(g0)), order='F') # Regularizor
w=np.ones(y.reshape(-1, 1).size).reshape((-1, 1)) # Weights
sM, tM = np.meshgrid(s, t, indexing='xy')
A = np.exp(-tM/sM)
g0 = g0 * y.sum() / (A # g0).sum() # Makes a "better guess" for the distribution, according to algorithm
print('msd of input data:\n', msd(g0, y, A, alpha, R, w, constraints))
for i in range(5): # Just for testing. If this is extremely high, ~1000, it's still bad.
g = fmin(func=msd,
x0 = g0,
args=(y, A, alpha, R, w, constraints),
**options,
disp=True)[:, np.newaxis]
msdfit = msd(g, y, A, alpha, R, w, constraints)
if 'zero_at_extremes' in constraints:
g[0] = 0
g[-1] = 0
if 'g>0' in constraints:
g = np.abs(g)
g0 = g
print('New guess', g)
print('Final msd of g', msdfit)
# Visualize the fit
plt.plot(s, g, label='Initial approximation')
plt.plot(np.logspace(-3, 6, 11), np.array([0, 0, 10.1625, 25.1974, 21.8711, 1.6377, 7.3895, 8.736, 1.4256, 0, 0]), label='Distribution to match')
plt.xscale('log')
plt.legend()
plt.show()
Matlab:
% Objective: match this distribution
g0 = [0 0 10.1625 25.1974 21.8711 1.6377 7.3895 8.736 1.4256 0 0]';
s0 = logspace(-3,6,length(g0))';
t = linspace(0.01,500,100)';
[sM,tM] = meshgrid(s0,t);
A = exp(-tM./sM);
rng(1);
% Creates data from the initial distribution with some random noise.
data = A*g0 + 0.07*rand(size(t));
% Parameters and function start
alpha = 1e-2; % regularization parameter
s = logspace(-3,6,20)'; % x of the ILT
g0 = ones(size(s)); % initial guess of y of ILT
y = data; % noisy data
options = optimset('MaxFunEvals',1e8,'MaxIter',1e8); % constraints for fminsearch
constraints = {'g>0','zero_at_the_extremes'}; % constraints for MSD
R = zeros(length(g0)-2,length(g0));
w = ones(size(y(:)));
[sM,tM] = meshgrid(s,t);
A = exp(-tM./sM);
g0 = g0*sum(y)/sum(A*g0); % Makes a "better guess" for the distribution
disp('msd of input data:')
disp(msd(g0, y, A, alpha, R, w, constraints))
for k = 1:5
[g,msdfit] = fminsearch(#msd,g0,options,y,A,alpha,R,w,constraints);
if ismember('zero_at_the_extremes',constraints)
g(1) = 0;
g(end) = 0;
end
if ismember('g>0',constraints)
g = abs(g);
end
g0 = g;
end
disp('New guess')
disp(g)
disp('Final msd of g')
disp(msdfit)
% Visualize the fit
semilogx(s, g)
hold on
semilogx(logspace(-3,6,11), [0 0 10.1625 25.1974 21.8711 1.6377 7.3895 8.736 1.4256 0 0])
legend('First approximation', 'Distribution to match')
hold off
function out = msd(g,y,A,alpha,R,w,constraints)
% msd: The mean square deviation; this is the function
% that has to be minimized by fminsearch
% Constraints and any 'a priori' knowledge
if ismember('zero_at_the_extremes',constraints)
g(1) = 0;
g(end) = 0;
end
if ismember('g>0',constraints)
g = abs(g); % must be g(i)>=0 for each i
end
r = diff(diff(g(1:end))); % second derivative of g
yfit = A*g;
% Sum of weighted square residuals
VAR = sum(w.*(y-yfit).^2);
% Regularizor
REG = alpha^2 * sum((r-R*g).^2);
% Output to be minimized
out = VAR+REG;
end
Here is the optimization in Python
Here is the optimization in Matlab
I have checked the output of MSD of g0 before starting, and both give the value of 2651. After minimization, Python goes up, to 4547, and Matlab goes down to 0.1381.
I think the problem is one of the following. It's in my implementation, that is, I am using fmin wrong, or there's some other passage I got wrong, but I can't figure out what. The fact the MSD increases when it should have decreased with a minimization function is damning. Reading the documentation, the scipy implementation is different from Matlab's (they use the Nelder Mead method described in Lagarias, per their documentation), while scipy uses the original Nelder Mead). Maybe that affects significantly? Or perhaps my initial guess is too bad for scipy's algorithm?
So, quite a long time since I posted this, but I wanted to share what I ended up learning and doing.
The Inverse Laplace Transform for CPMG data is a bit of a misnomer, and it's more properly called just inversion. The general problem is solving a Fredholm integral of the first kind. One way of doing this is the Tikhonov regularization method. Turns out, you can describe this problem quite easily using numpy, and solve it with a scipy package, so I don't have to "reinvent" the wheel with this.
I used the solution shown in this post, and the names here reflect that solution.
def tikhonov_regularized_inversion(
kernel: np.ndarray, alpha: float, data: np.ndarray
) -> np.ndarray:
data = data.reshape(-1, 1)
I = alpha * np.eye(*kernel.shape)
C = np.concatenate([kernel, I], axis=0)
d = np.concatenate([data, np.zeros_like(data)])
x, _ = nnls(C, d.flatten())
Here, kernel is a matrix containing all the possible exponential decay curves, and my solution judges the contribution of each decay curve in the data I received. First, I stack my data as a column, then pad it with zeros, creating the vector d. I then stack my kernel on top of a diagonal matrix containing the regularization parameter alpha along the diagonal, of the same size as the kernel. Last, I call the convenient nnls, a non negative least square solver in scipy.optimize. This is because there's no reason to have a negative contribution, only no contribution.
This solved my problem, it's quick and convenient.
I'm trying to solve an integral equation using the following code (irrelevant parts removed):
def _pdf(self, a, b, c, t):
pdf = some_pdf(a,b,c,t)
return pdf
def _result(self, a, b, c, flag):
return fsolve(lambda t: flag - 1 + quad(lambda tau: self._pdf(a, b, c, tau), 0, t)[0], x0)[0]
Which takes a probability density function and finds a result tau such that the integral of pdf from tau to infinity is equal to flag. Note that x0 is a (float) estimate of the root defined elsewhere in the script. Also note that flag is an extremely small number, on the order of 1e-9.
In my application fsolve only successfully finds a root about 50% of the time. It often just returns x0, significantly biasing my results. There is no closed form for the integral of pdf, so I am forced to integrate numerically and feel that this might be introducing some inaccuracy?
EDIT:
This has since been solved using a method other than that described below, but I'd like to get quadpy to work and see if the results improve at all. The specific code I'm trying to get to work is as follows:
import quadpy
import numpy as np
from scipy.optimize import *
from scipy.special import gammaln, kv, gammaincinv, gamma
from scipy.integrate import quad, simps
l = 226.02453163
mu = 0.00212571582056
nu = 4.86569872444
flag = 2.5e-09
estimate = 3 * mu
def pdf(l, mu, nu, t):
return np.exp(np.log(2) + (l + nu - 1 + 1) / 2 * np.log(l * nu / mu) + (l + nu - 1 - 1) / 2 * np.log(t) + np.log(
kv(nu - l, 2 * np.sqrt(l * nu / mu * t))) - gammaln(l) - gammaln(nu))
def tail_cdf(l, mu, nu, tau):
i, error = quadpy.line_segment.adaptive_integrate(
lambda t: pdf(l, mu, nu, t), [tau, 10000], 1.0e-10
)
return i
result = fsolve(lambda tau: flag - tail_cdf(l, mu, nu, tau[0]), estimate)
When I run this I get an assertion error from assert all(lengths > minimum_interval_length). I'm not quite sure of how to remedy this; any help would be very much appreciated!
As an example, I tried 1 / x for the integration between 1 and alpha to retrieve the target integral 2.0. This
import quadpy
from scipy.optimize import fsolve
def f(alpha):
beta, _ = quadpy.quad(lambda x: 1.0/x, 1, alpha)
return beta
target = 2.0
res = fsolve(lambda alpha: target - f(alpha), x0=2.0)
print(res)
correctly returns 7.38905611.
The failing quadpy assertion
assert all(lengths > minimum_interval_length)
you're getting means that the adaptive integration hit its limit: Either relax your tolerance a bit, or decrease the minimum_interval_length (see here).
I'm looking for a way to plot a curve through some experimental data. The data shows a small linear regime with a shallow gradient, followed by a steep linear regime after a threshold value.
My data is here: http://pastebin.com/H4NSbxqr
I could fit the data with two lines relatively easily, but I'd like to fit with a continuous line ideally - which should look like two lines with a smooth curve joining them around the threshold (~5000 in the data, shown above).
I attempted this using scipy.optimize curve_fit and trying a function which included the sum of a straight line and an exponential:
y = a*x + b + c*np.exp((x-d)/e)
although despite numerous attempts, it didn't find a solution.
If anyone has any suggestions please, either on the choice of fitting distribution / method or the curve_fit implementation, they would be greatly appreciated.
If you don't have a particular reason to believe that linear + exponential is the true underlying cause of your data, then I think a fit to two lines makes the most sense. You can do this by making your fitting function the maximum of two lines, for example:
import numpy as np
import matplotlib.pyplot as plt
from scipy.optimize import curve_fit
def two_lines(x, a, b, c, d):
one = a*x + b
two = c*x + d
return np.maximum(one, two)
Then,
x, y = np.genfromtxt('tmp.txt', unpack=True, delimiter=',')
pw0 = (.02, 30, .2, -2000) # a guess for slope, intercept, slope, intercept
pw, cov = curve_fit(two_lines, x, y, pw0)
crossover = (pw[3] - pw[1]) / (pw[0] - pw[2])
plt.plot(x, y, 'o', x, two_lines(x, *pw), '-')
If you really want a continuous and differentiable solution, it occurred to me that a hyperbola has a sharp bend to it, but it has to be rotated. It was a bit difficult to implement (maybe there's an easier way), but here's a go:
def hyperbola(x, a, b, c, d, e):
""" hyperbola(x) with parameters
a/b = asymptotic slope
c = curvature at vertex
d = offset to vertex
e = vertical offset
"""
return a*np.sqrt((b*c)**2 + (x-d)**2)/b + e
def rot_hyperbola(x, a, b, c, d, e, th):
pars = a, b, c, 0, 0 # do the shifting after rotation
xd = x - d
hsin = hyperbola(xd, *pars)*np.sin(th)
xcos = xd*np.cos(th)
return e + hyperbola(xcos - hsin, *pars)*np.cos(th) + xcos - hsin
Run it as
h0 = 1.1, 1, 0, 5000, 100, .5
h, hcov = curve_fit(rot_hyperbola, x, y, h0)
plt.plot(x, y, 'o', x, two_lines(x, *pw), '-', x, rot_hyperbola(x, *h), '-')
plt.legend(['data', 'piecewise linear', 'rotated hyperbola'], loc='upper left')
plt.show()
I was also able to get the line + exponential to converge, but it looks terrible. This is because it's not a good descriptor of your data, which is linear and an exponential is very far from linear!
def line_exp(x, a, b, c, d, e):
return a*x + b + c*np.exp((x-d)/e)
e0 = .1, 20., .01, 1000., 2000.
e, ecov = curve_fit(line_exp, x, y, e0)
If you want to keep it simple, there's always a polynomial or spline (piecewise polynomials)
from scipy.interpolate import UnivariateSpline
s = UnivariateSpline(x, y, s=x.size) #larger s-value has fewer "knots"
plt.plot(x, s(x))
I researched this a little, Applied Linear Regression by Sanford, and the Correlation and Regression lecture by Steiger had some good info on it. They all however lack the right model, the piecewise function should be
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import lmfit
dfseg = pd.read_csv('segreg.csv')
def err(w):
th0 = w['th0'].value
th1 = w['th1'].value
th2 = w['th2'].value
gamma = w['gamma'].value
fit = th0 + th1*dfseg.Temp + th2*np.maximum(0,dfseg.Temp-gamma)
return fit-dfseg.C
p = lmfit.Parameters()
p.add_many(('th0', 0.), ('th1', 0.0),('th2', 0.0),('gamma', 40.))
mi = lmfit.minimize(err, p)
lmfit.printfuncs.report_fit(mi.params)
b0 = mi.params['th0']; b1=mi.params['th1'];b2=mi.params['th2']
gamma = int(mi.params['gamma'].value)
import statsmodels.formula.api as smf
reslin = smf.ols('C ~ 1 + Temp + I((Temp-%d)*(Temp>%d))' % (gamma,gamma), data=dfseg).fit()
print reslin.summary()
x0 = np.array(range(0,gamma,1))
x1 = np.array(range(0,80-gamma,1))
y0 = b0 + b1*x0
y1 = (b0 + b1 * float(gamma) + (b1 + b2)* x1)
plt.scatter(dfseg.Temp, dfseg.C)
plt.hold(True)
plt.plot(x0,y0)
plt.plot(x1+gamma,y1)
plt.show()
Result
[[Variables]]
th0: 78.6554456 +/- 3.966238 (5.04%) (init= 0)
th1: -0.15728297 +/- 0.148250 (94.26%) (init= 0)
th2: 0.72471237 +/- 0.179052 (24.71%) (init= 0)
gamma: 38.3110177 +/- 4.845767 (12.65%) (init= 40)
The data
"","Temp","C"
"1",8.5536,86.2143
"2",10.6613,72.3871
"3",12.4516,74.0968
"4",16.9032,68.2258
"5",20.5161,72.3548
"6",21.1613,76.4839
"7",24.3929,83.6429
"8",26.4839,74.1935
"9",26.5645,71.2581
"10",27.9828,78.2069
"11",32.6833,79.0667
"12",33.0806,71.0968
"13",33.7097,76.6452
"14",34.2903,74.4516
"15",36,56.9677
"16",37.4167,79.8333
"17",43.9516,79.7097
"18",45.2667,76.9667
"19",47,76
"20",47.1129,78.0323
"21",47.3833,79.8333
"22",48.0968,73.9032
"23",49.05,78.1667
"24",57.5,81.7097
"25",59.2,80.3
"26",61.3226,75
"27",61.9194,87.0323
"28",62.3833,89.8
"29",64.3667,96.4
"30",65.371,88.9677
"31",68.35,91.3333
"32",70.7581,91.8387
"33",71.129,90.9355
"34",72.2419,93.4516
"35",72.85,97.8333
"36",73.9194,92.4839
"37",74.4167,96.1333
"38",76.3871,89.8387
"39",78.0484,89.4516
Graph
I used #user423805 's answer (found via google groups thread: https://groups.google.com/forum/#!topic/lmfit-py/7I2zv2WwFLU ) but noticed it had some limitations when trying to use three or more segments.
Instead of applying np.maximum in the minimizer error function or adding (b1 + b2) in #user423805 's answer, I used the same linear spline calculation for both the minimizer and end-usage:
# least_splines_calc works like this for an example with three segments
# (four threshold params, three gamma params):
#
# for 0 < x < gamma0 : y = th0 + (th1 * x)
# for gamma0 < x < gamma1 : y = th0 + (th1 * x) + (th2 * (x - gamma0))
# for gamma1 < x : y = th0 + (th1 * x) + (th2 * (x - gamma0)) + (th3 * (x - gamma1))
#
def least_splines_calc(x, thresholds, gammas):
if(len(thresholds) < 2):
print("Error: expected at least two thresholds")
return None
applicable_gammas = filter(lambda gamma: x > gamma , gammas)
#base result
y = thresholds[0] + (thresholds[1] * x)
#additional factors calculated depending on x value
for i in range(0, len(applicable_gammas)):
y = y + ( thresholds[i + 2] * ( x - applicable_gammas[i] ) )
return y
def least_splines_calc_array(x_array, thresholds, gammas):
y_array = map(lambda x: least_splines_calc(x, thresholds, gammas), x_array)
return y_array
def err(params, x, data):
th0 = params['th0'].value
th1 = params['th1'].value
th2 = params['th2'].value
th3 = params['th3'].value
gamma1 = params['gamma1'].value
gamma2 = params['gamma2'].value
thresholds = np.array([th0, th1, th2, th3])
gammas = np.array([gamma1, gamma2])
fit = least_splines_calc_array(x, thresholds, gammas)
return np.array(fit)-np.array(data)
p = lmfit.Parameters()
p.add_many(('th0', 0.), ('th1', 0.0),('th2', 0.0),('th3', 0.0),('gamma1', 9.),('gamma2', 9.3)) #NOTE: the 9. / 9.3 were guesses specific to my data, you will need to change these
mi = lmfit.minimize(err_alt, p, args=(np.array(dfseg.Temp), np.array(dfseg.C)))
After minimization, convert the params found by the minimizer into an array of thresholds and gammas to re-use linear_splines_calc to plot the linear splines regression.
Reference: While there's various places that explain least splines (I think #user423805 used http://www.statpower.net/Content/313/Lecture%20Notes/Splines.pdf , which has the (b1 + b2) addition I disagree with in its sample code despite similar equations) , the one that made the most sense to me was this one (by Rob Schapire / Zia Khan at Princeton) : https://www.cs.princeton.edu/courses/archive/spring07/cos424/scribe_notes/0403.pdf - section 2.2 goes into linear splines. Excerpt below:
If you're looking to join what appears to be two straight lines with a hyperbola having a variable radius at/near the intersection of the two lines (which are its asymptotes), I urge you to look hard at Using an Hyperbola as a Transition Model to Fit Two-Regime Straight-Line Data, by Donald G. Watts and David W. Bacon, Technometrics, Vol. 16, No. 3 (Aug., 1974), pp. 369-373.
The formula is drop dead simple, nicely adjustable, and works like a charm. From their paper (in case you can't access it):
As a more useful alternative form we consider an hyperbola for which:
(i) the dependent variable y is a single valued function of the independent variable x,
(ii) the left asymptote has slope theta_1,
(iii) the right asymptote has slope theta_2,
(iv) the asymptotes intersect at the point (x_o, beta_o),
(v) the radius of curvature at x = x_o is proportional to a quantity delta. Such an hyperbola can be written y = beta_o + beta_1*(x - x_o) + beta_2* SQRT[(x - x_o)^2 + delta^2/4], where beta_1 = (theta_1 + theta_2)/2 and beta_2 = (theta_2 - theta_1)/2.
delta is the adjustable parameter that allows you to either closely follow the lines right to the intersection point or smoothly merge from one line to the other.
Just solve for the intersection point (x_o, beta_o), and plug into the formula above.
BTW, in general, if line 1 is y_1 = b_1 + m_1 *x and line 2 is y_2 = b_2 + m_2 * x, then they intersect at x* = (b_2 - b_1) / (m_1 - m_2) and y* = b_1 + m_1 * x*. So, to connect with the formalism above, x_o = x*, beta_o = y* and the two m_*'s are the two thetas.
There is a straightforward method (not iterative, no initial guess) pp.12-13 in https://fr.scribd.com/document/380941024/Regression-par-morceaux-Piecewise-Regression-pdf
The data comes from the scanning of the figure published by IanRoberts in his question. Scanning for the coordinates of the pixels in not accurate. So, don't be surprised by additional deviation.
Note that the abscisses and ordinates scales have been devised by 1000.
The equations of the two segments are
The approximate values of the five parameters are written on the above figure.
I have another question that I was hoping someone could help me with.
I'm using the Jensen-Shannon-Divergence to measure the similarity between two probability distributions. The similarity scores appear to be correct in the sense that they fall between 1 and 0 given that one uses the base 2 logarithm, with 0 meaning that the distributions are equal.
However, I'm not sure whether there is in fact an error somewhere and was wondering whether someone might be able to say 'yes it's correct' or 'no, you did something wrong'.
Here is the code:
from numpy import zeros, array
from math import sqrt, log
class JSD(object):
def __init__(self):
self.log2 = log(2)
def KL_divergence(self, p, q):
""" Compute KL divergence of two vectors, K(p || q)."""
return sum(p[x] * log((p[x]) / (q[x])) for x in range(len(p)) if p[x] != 0.0 or p[x] != 0)
def Jensen_Shannon_divergence(self, p, q):
""" Returns the Jensen-Shannon divergence. """
self.JSD = 0.0
weight = 0.5
average = zeros(len(p)) #Average
for x in range(len(p)):
average[x] = weight * p[x] + (1 - weight) * q[x]
self.JSD = (weight * self.KL_divergence(array(p), average)) + ((1 - weight) * self.KL_divergence(array(q), average))
return 1-(self.JSD/sqrt(2 * self.log2))
if __name__ == '__main__':
J = JSD()
p = [1.0/10, 9.0/10, 0]
q = [0, 1.0/10, 9.0/10]
print J.Jensen_Shannon_divergence(p, q)
The problem is that I feel that the scores are not high enough when comparing two text documents, for instance. However, this is purely a subjective feeling.
Any help is, as always, appreciated.
Note that the scipy entropy call below is the Kullback-Leibler divergence.
See: http://en.wikipedia.org/wiki/Jensen%E2%80%93Shannon_divergence
#!/usr/bin/env python
from scipy.stats import entropy
from numpy.linalg import norm
import numpy as np
def JSD(P, Q):
_P = P / norm(P, ord=1)
_Q = Q / norm(Q, ord=1)
_M = 0.5 * (_P + _Q)
return 0.5 * (entropy(_P, _M) + entropy(_Q, _M))
Also note that the test case in the Question looks erred?? The sum of the p distribution does not add to 1.0.
See: http://www.itl.nist.gov/div898/handbook/eda/section3/eda361.htm
Since the Jensen-Shannon distance (distance.jensenshannon) has been included in Scipy 1.2, the Jensen-Shannon divergence can be obtained as the square of the Jensen-Shannon distance:
from scipy.spatial import distance
distance.jensenshannon([1.0/10, 9.0/10, 0], [0, 1.0/10, 9.0/10]) ** 2
# 0.5306056938642212
Get some data for distributions with known divergence and compare your results against those known values.
BTW: the sum in KL_divergence may be rewritten using the zip built-in function like this:
sum(_p * log(_p / _q) for _p, _q in zip(p, q) if _p != 0)
This does away with lots of "noise" and is also much more "pythonic". The double comparison with 0.0 and 0 is not necessary.
A general version, for n probability distributions, in python
import numpy as np
from scipy.stats import entropy as H
def JSD(prob_distributions, weights, logbase=2):
# left term: entropy of misture
wprobs = weights * prob_distributions
mixture = wprobs.sum(axis=0)
entropy_of_mixture = H(mixture, base=logbase)
# right term: sum of entropies
entropies = np.array([H(P_i, base=logbase) for P_i in prob_distributions])
wentropies = weights * entropies
sum_of_entropies = wentropies.sum()
divergence = entropy_of_mixture - sum_of_entropies
return(divergence)
# From the original example with three distributions:
P_1 = np.array([1/2, 1/2, 0])
P_2 = np.array([0, 1/10, 9/10])
P_3 = np.array([1/3, 1/3, 1/3])
prob_distributions = np.array([P_1, P_2, P_3])
n = len(prob_distributions)
weights = np.empty(n)
weights.fill(1/n)
print(JSD(prob_distributions, weights))
#0.546621319446
Explicitly following the math in the Wikipedia article:
def jsdiv(P, Q):
"""Compute the Jensen-Shannon divergence between two probability distributions.
Input
-----
P, Q : array-like
Probability distributions of equal length that sum to 1
"""
def _kldiv(A, B):
return np.sum([v for v in A * np.log2(A/B) if not np.isnan(v)])
P = np.array(P)
Q = np.array(Q)
M = 0.5 * (P + Q)
return 0.5 * (_kldiv(P, M) +_kldiv(Q, M))