I have a set of coordinates (x, y, z(x, y)) which describe intensities (z) at coordinates x, y. For a set number of these intensities at different coordinates, I need to fit a 2D Gaussian that minimizes the mean squared error.
The data is in numpy matrices and for each fitting session I will have either 4, 9, 16 or 25 coordinates. Ultimately I just need to get the central position of the gaussian (x_0, y_0) that has smallest MSE.
All of the examples that I have found use scipy.optimize.curve_fit but the input data they have is over an entire mesh rather than a few coordinates.
Any help would be appreciated.
Introduction
There are multiple ways to approach this. You can use non-linear methods (e.g. scipy.optimize.curve_fit), but they'll be slow and aren't guaranteed to converge. You can linearize the problem (fast, unique solution), but any noise in the "tails" of the distribution will cause issues. There are actually a few tricks you can apply to this particular case to avoid the latter issue. I'll show some examples, but I don't have time to demonstrate all of the "tricks" right now.
Just as a side note, a general 2D guassian has 6 parameters, so you won't be able to fully fit things with 4 points. However, it sounds like you might be assuming that there's no covariance between x and y and that the variances are the same in each direction (i.e. a perfectly "round" bell curve). If that's the case, then you only need four parameters. If you know the amplitude of the guassian, you'll only need three. However, I'm going to start with the general solution, and you can simplify it later on, if you want to.
For the moment, let's focus on solving this problem using non-linear methods (e.g. scipy.optimize.curve_fit).
The general equation for a 2D guassian is (directly from wikipedia):
where:
is essentially 0.5 over the covariance matrix, A is the amplitude,
and (X₀, Y₀) is the center
Generate simplified sample data
Let's write the equation above out:
import numpy as np
import matplotlib.pyplot as plt
def gauss2d(x, y, amp, x0, y0, a, b, c):
inner = a * (x - x0)**2
inner += 2 * b * (x - x0)**2 * (y - y0)**2
inner += c * (y - y0)**2
return amp * np.exp(-inner)
And then let's generate some example data. To start with, we'll generate some data that will be easy to fit:
np.random.seed(1977) # For consistency
x, y = np.random.random((2, 10))
x0, y0 = 0.3, 0.7
amp, a, b, c = 1, 2, 3, 4
zobs = gauss2d(x, y, amp, x0, y0, a, b, c)
fig, ax = plt.subplots()
scat = ax.scatter(x, y, c=zobs, s=200)
fig.colorbar(scat)
plt.show()
Note that we haven't added any noise, and the center of the distribution is within the range that we have data (i.e. center at 0.3, 0.7 and a scatter of x,y observations between 0 and 1). For the moment, let's stick with this, and then we'll see what happens when we add noise and shift the center.
Non-linear fitting
To start with, let's use scpy.optimize.curve_fit to preform a non-linear least-squares fit to the gaussian function. (On a side note, you can play around with the exact minimization algorithm by using some of the other functions in scipy.optimize.)
The scipy.optimize functions expect a slightly different function signature than the one we originally wrote above. We could write a wrapper to "translate", but let's just re-write the gauss2d function instead:
def gauss2d(xy, amp, x0, y0, a, b, c):
x, y = xy
inner = a * (x - x0)**2
inner += 2 * b * (x - x0)**2 * (y - y0)**2
inner += c * (y - y0)**2
return amp * np.exp(-inner)
All we did was have the function expect the independent variables (x & y) as a single 2xN array.
Now we need to make an initial guess at what the guassian curve's parameters actually are. This is optional (the default is all ones, if I recall correctly), but you're likely to have problems converging if 1, 1 is not particularly close to the "true" center of the gaussian curve. For that reason, we'll use the x and y values of our largest observed z-value as a starting point for the center. I'll leave the rest of the parameters as 1, but if you know that they're likely to consistently be significantly different, change them to something more reasonable.
Here's the full, stand-alone example:
import numpy as np
import scipy.optimize as opt
import matplotlib.pyplot as plt
def main():
x0, y0 = 0.3, 0.7
amp, a, b, c = 1, 2, 3, 4
true_params = [amp, x0, y0, a, b, c]
xy, zobs = generate_example_data(10, true_params)
x, y = xy
i = zobs.argmax()
guess = [1, x[i], y[i], 1, 1, 1]
pred_params, uncert_cov = opt.curve_fit(gauss2d, xy, zobs, p0=guess)
zpred = gauss2d(xy, *pred_params)
print 'True parameters: ', true_params
print 'Predicted params:', pred_params
print 'Residual, RMS(obs - pred):', np.sqrt(np.mean((zobs - zpred)**2))
plot(xy, zobs, pred_params)
plt.show()
def gauss2d(xy, amp, x0, y0, a, b, c):
x, y = xy
inner = a * (x - x0)**2
inner += 2 * b * (x - x0)**2 * (y - y0)**2
inner += c * (y - y0)**2
return amp * np.exp(-inner)
def generate_example_data(num, params):
np.random.seed(1977) # For consistency
xy = np.random.random((2, num))
zobs = gauss2d(xy, *params)
return xy, zobs
def plot(xy, zobs, pred_params):
x, y = xy
yi, xi = np.mgrid[:1:30j, -.2:1.2:30j]
xyi = np.vstack([xi.ravel(), yi.ravel()])
zpred = gauss2d(xyi, *pred_params)
zpred.shape = xi.shape
fig, ax = plt.subplots()
ax.scatter(x, y, c=zobs, s=200, vmin=zpred.min(), vmax=zpred.max())
im = ax.imshow(zpred, extent=[xi.min(), xi.max(), yi.max(), yi.min()],
aspect='auto')
fig.colorbar(im)
ax.invert_yaxis()
return fig
main()
In this case, we exactly(ish) recover our original "true" parameters.
True parameters: [1, 0.3, 0.7, 2, 3, 4]
Predicted params: [ 1. 0.3 0.7 2. 3. 4. ]
Residual, RMS(obs - pred): 1.01560615193e-16
As we'll see in a second, this won't always be the case...
Adding Noise
Let's add some noise to our observations. All I've done here is change the generate_example_data function:
def generate_example_data(num, params):
np.random.seed(1977) # For consistency
xy = np.random.random((2, num))
noise = np.random.normal(0, 0.3, num)
zobs = gauss2d(xy, *params) + noise
return xy, zobs
However, the result looks quite different:
And as far as the parameters go:
True parameters: [1, 0.3, 0.7, 2, 3, 4]
Predicted params: [ 1.129 0.263 0.750 1.280 32.333 10.103 ]
Residual, RMS(obs - pred): 0.152444640098
The predicted center hasn't changed much, but the b and c parameters have changed quite a bit.
If we change the center of the function to somewhere slightly outside of our scatter of points:
x0, y0 = -0.3, 1.1
We'll wind up with complete nonsense as a result in the presence of noise! (It still works correctly without noise.)
True parameters: [1, -0.3, 1.1, 2, 3, 4]
Predicted params: [ 0.546 -0.939 0.857 -0.488 44.069 -4.136]
Residual, RMS(obs - pred): 0.235664449826
This is a common problem when fitting a function that decays to zero. Any noise in the "tails" can result in a very poor result. There are a number of strategies to deal with this. One of the easiest is to weight the inversion by the observed z-values. Here's an example for the 1D case: (focusing on linearized the problem) How can I perform a least-squares fitting over multiple data sets fast? If I have time later, I'll add an example of this for the 2D case.
Related
I display a gyroid structure (TPMS) in a cartesian system using Pyvista. I try now to display the structure in cylindrical coordinates. Pyvista displays something cylindrical indeed but it seems that the unit cell length is not uniform (while there is no reason to change this my parameter "a" being steady. This change seems to appear especially along z but I don't understand why (see image).
I have this:
Here is a part of my code.
Thank you for your help.
import pyvista as pv
import numpy as np
from numpy import cos, sin, pi
from random import uniform
lattice_par = 1.0 # Unit cell length
a = (2*pi)/lattice_par
res = 200j
r, theta, z = np.mgrid[0:2:res, 0:2*pi:res, 0:4:res]
# consider using non-equidistant r for uniformity
def GyroidCyl(r, theta, z, b=0.8):
return (sin(a*(r*cos(theta) - 1))*cos(a*(r*sin(theta) - 1))
+ sin(a*(r*sin(theta) - 1))*cos(a*(z - 1))
+ sin(a*(z - 1))*cos(a*(r*cos(theta) - 1))
- b)
vol3 = GyroidCyl(r, theta, z)
# compute Cartesian coordinates for grid points
x = r * cos(theta)
y = r * sin(theta)
grid = pv.StructuredGrid(x, y, z)
grid["vol3"] = vol3.flatten()
contours3 = grid.contour([0]) # Isosurface = 0
pv.set_plot_theme('document')
p = pv.Plotter()
p.add_mesh(contours3, scalars=contours3.points[:, 2], show_scalar_bar=False, interpolate_before_map=True,
show_edges=False, smooth_shading=False, render=True)
p.show_axes_all()
p.add_floor()
p.show_grid()
p.add_title('Gyroid in cylindrical coordinates')
p.add_text('Volume Fraction Parameter = ' + str(b))
p.show(window_size=[2040, 1500])
So you've noted in comments that you're trying to replicate something like the strategy explained in this paper. What they do is take a regular gyroid unit cell, and then transform it to build a cylindrical shell. If igloos were cylindrical, then a gyroid cell would be a single piece of snow brick. Put them next to one another and stack them in a column, and you get a cylinder.
Since I can't use figures from the paper we'll have to recreate some ourselves. So you have to start from a regular gyroid defined by the implicit function
cos(x) sin(y) + cos(y) sin(z) + cos(z) sin(x) = 0
(or some variation thereof). Here's how a single unit cell looks:
import pyvista as pv
import numpy as np
res = 100j
a = 2*np.pi
x, y, z = np.mgrid[0:a:res, 0:a:res, 0:a:res]
def Gyroid(x, y, z):
return np.cos(x)*np.sin(y) + np.cos(y)*np.sin(z) + np.cos(z)*np.sin(x)
# compute implicit function
fun_values = Gyroid(x, y, z)
# create grid for contouring
grid = pv.StructuredGrid(x, y, z)
grid["vol3"] = fun_values.ravel('F')
contours3 = grid.contour([0]) # isosurface for 0
# plot the contour, i.e. the gyroid
pv.set_plot_theme('document')
plotter = pv.Plotter()
plotter.add_mesh(contours3, scalars=contours3.points[:, -1],
show_scalar_bar=False)
plotter.add_bounding_box()
plotter.enable_terrain_style()
plotter.show_axes()
plotter.show()
Using the "unit cell" term implies there's an underlying infinite lattice, which can be built by stacking these (rectangular) unit cells neatly next to one another. With some imagination we can convince ourselves that this is true. Or we can look at the formula and note that due to the trigonometric functions the function is periodic in x, y and z, with period 2*pi. This also tells us that we can transform the unit cell to have arbitrary rectangular dimensions by introducing lattice parameters a, b and c:
cos(kx x) sin(ky y) + cos(ky y) sin(kz z) + cos(kz z) sin(kx x) = 0, where
kx = 2 pi/a
ky = 2 pi/b
kz = 2 pi/c
(These kx, ky and kz quantities are called wave vectors in solid state physics.)
The corresponding change only affects the header:
res = 100j
a, b, c = lattice_params = 1, 2, 3
kx, ky, kz = [2*np.pi/lattice_param for lattice_param in lattice_params]
x, y, z = np.mgrid[0:a:res, 0:b:res, 0:c:res]
def Gyroid(x, y, z):
return ( np.cos(kx*x)*np.sin(ky*y)
+ np.cos(ky*y)*np.sin(kz*z)
+ np.cos(kz*z)*np.sin(kx*x))
This is where we start. What we have to do is take this unit cell, bend it so that it corresponds to a 30-degree circular arc on a cylinder, and stack the cylinder using this unit. According to the paper, they used 12 unit cells to create a circle in a plane (hence the 30-degree magic number), and stacked three such circular bands on top of each other to build the cylinder.
The actual mapping is also fairly clearly explained in the paper. Whereas your original x, y and z parameters of the function essentially interpolated between [0, a], [0, b] and [0, c], respectively, in the new setup x interpolates in the radius range [r1, r2], y interpolates in the angular range [0, pi/6] and z is just z. (In the paper x and y seem to be reversed with respect to this convention, but this shouldn't matter. If it matters, that's left as an exercise to the reader.)
So what we need to do is more or less keep the current grid points, but transform the corresponding x, y and z grid points so that they lie on a cylinder instead. Here's one take:
import pyvista as pv
import numpy as np
res = 100j
a, b, c = lattice_params = 1, 1, 1
kx, ky, kz = [2*np.pi/lattice_param for lattice_param in lattice_params]
r_aux, phi, z = np.mgrid[0:a:res, 0:b:res, 0:3*c:res]
# convert r_aux range to actual radii
r1, r2 = 1.5, 2
r = r2/a*r_aux + r1/a*(1 - r_aux)
def Gyroid(x, y, z):
return ( np.cos(kx*x)*np.sin(ky*y)
+ np.cos(ky*y)*np.sin(kz*z)
+ np.cos(kz*z)*np.sin(kx*x))
# compute data for cylindrical gyroid
# r_aux is x, phi / 12 is y and z is z
fun_values = Gyroid(r_aux, phi * 12, z)
# compute Cartesian coordinates for grid points
x = r * np.cos(phi*ky)
y = r * np.sin(phi*ky)
grid = pv.StructuredGrid(x, y, z)
grid["vol3"] = fun_values.ravel('F')
contours3 = grid.contour([0])
# plot cylindrical gyroid
pv.set_plot_theme('document')
plotter = pv.Plotter()
plotter.add_mesh(contours3, scalars=contours3.points[:, -1],
show_scalar_bar=False)
plotter.add_bounding_box()
plotter.show_axes()
plotter.enable_terrain_style()
plotter.show()
If you want to look at a single transformed unit cell in the cylindrical setting, use a single domain of phi and z for the function and only convert to 1/12 a full circle for the grid points:
fun_values = Gyroid(r_aux, phi, z/3)
# compute Cartesian coordinates for grid points
x = r * np.cos(phi*ky/12)
y = r * np.sin(phi*ky/12)
grid = pv.StructuredGrid(x, y, z/3)
But it's not easy to see the curvature in the (no longer a) unit cell:
I counted the integral and I want to display it on the graph, but I wonder how it should be correctly placed on the graph. It seems to me that plt.plot() alone is not enough, or maybe I am wrong, I would like to know the correct way to display this result in a graph.
import matplotlib.pyplot as plt
import numpy as np
from scipy.integrate import quad
def integral(x, a, b):
return a * np.log(x + b)
a = 3
b = 2
I = quad(integral, 1, 5, args=(a, b))
print(I)
plt.plot()
plt.show()
I assume you know calculus but not so much about programming.
matplotlib.plot only plots data, so you have to construct an array with the datapoints you want to plot. Also the result of quad is a pair of numbers, the definite integral approximation and an estimated bound for the numerical errors.
If you want to plot the antiderivative of a function you will have to compute the integral for each of the points you want to display.
Here is an example in which I create an array and compute the integral between each element a[i] < x < a[i+1], and use a cumulative sum to get the curve.
For reference I also plotted the analytic integral
import matplotlib.pyplot as plt
import numpy as np
from scipy.integrate import quad
def integral(x, a, b):
return a * np.log(x + b)
def II(a, b, x0, x1 = None):
if x1 is None:
# indefinite integral
return a * ((x0 + b) * np.log(x0 + b) - x0)
else:
# definite integral
return II(a,b, x1) - II(a,b, x0)
a = 3
b = 2
# plot 100 points equally spaced for 1 < x < 5
x = np.linspace(1, 5, 100)
# The first column of I is the value of the integral, the second is the accumulated error
# I[i,0] is the integral from x[0] to x[i+1].
I = np.cumsum([quad(integral, x[i], x[i+1], args=(a, b)) for i in range(len(x) - 1)], axis=0);
# Now you can plot I
plt.plot(x[1:], I[:,0])
plt.plot(x[1:], II(a, b, x[0], x[1:]), '--r')
plt.show()
Also take the tour if you didn't already, to know how to classify the answers you receive.
I hope this answers your question.
I am trying to fit a gaussian to my data which is taken in a pretty narrow spectral window. We got about 2 points of continuum and then about 10-11 that are part of the line. It should still be possible to fit it I think, but the curve fit is failing each time, and I am not sure why.
When running I get RuntimeError: Optimal parameters not found: Number of calls to function has reached maxfev = 800.
Code and data:
import matplotlib.pyplot as plt
from scipy.optimize import curve_fit
import numpy as np
x = np.arange(13)
xx = np.arange(130)/13.
y = np.array([19699.959 , 21679.445 , 21143.195 , 20602.875 , 16246.769 ,
11635.25 , 8602.465 , 7035.493 , 6697.0337, 6510.092 ,
7717.772 , 12270.446 , 16807.81 ])
# weighted arithmetic mean (corrected - check the section below)
mean = sum(x * y) / sum(y)
sigma = np.sqrt(sum(y * (x - mean)**2) / sum(y))
def Gauss(x, a, x0, sigma):
return a * np.exp(-(x - x0)**2 / (2 * sigma**2))
popt,pcov = curve_fit(Gauss, x, y, p0=[max(y), mean, sigma])
plt.plot(x, y, 'b+:', label='data')
plt.plot(xx, Gauss(xx, *popt), 'r-', label='fit')
plt.legend()
plt.show()
As the error says, the procedure to find the optimal values doesn't converge. If you really think that what you have can be fitted with a Gaussian curve, then this in general means you have a bad starting point.
How you're giving the starting point could have been an issue, particularly with how you provide sigma given that at positions 11, 12 and 13 you have what could be the onset of another signal. Anyhow, that's not the biggest issue this time, but the fact that you forgot to add an offset to the gaussian function
# ----> new parameter in signature
# |
def Gauss(x, y0, a, x0, sigma):
return y0 + a * np.exp(-(x - x0)**2 / (2 * sigma**2))
# |
# -------> adding and offset
Then, you can decide how to provide a starting point for the offset, but by eye, I did set 5000
popt, pcov = curve_fit(Gauss, x, y, p0=[5000, max(y), mean, sigma])
Doing that, I get a fit. But, due to the last three data points, it's not a very nice one.
If you avoid those values, the fit improves significantly.
Edit:
As indicated in the comments, the Gaussian is centered at about 8 looking downwards (silly me, it was an absorption line).
In such a case, the offset should be located at about the maximum ~22000 and then the parameter for the amplitude should be negative ~ -(max(y)-min(y)) ~ -16000.
And as an addition, better change xx to be as follows
xx = np.linspace(0, 13, 100)
or
xx = np.arange(0, 13, 0.05)
Which will give
and checking popt you get basically the values I mentioned/estimated by just looking at the plot ~(2180, -16000, 8) with a sigma of 2.7 which was the only one I don't have an immediate feeling on how to estimate.
My guess is that you should actually be fitting a mixture of Gauss and a Cauchy/Lorentz lineshape or even better a Voigt lineshape, to account for experimental broadening.
Suppose I have x and y vectors with a weight vector wgt. I can fit a cubic curve (y = a x^3 + b x^2 + c x + d) by using np.polyfit as follows:
y_fit = np.polyfit(x, y, deg=3, w=wgt)
Now, suppose I want to do another fit, but this time, I want the fit to pass through 0 (i.e. y = a x^3 + b x^2 + c x, d = 0), how can I specify a particular coefficient (i.e. d in this case) to be zero?
Thanks
You can try something like the following:
Import curve_fit from scipy, i.e.
from scipy.optimize import curve_fit
import matplotlib.pyplot as plt
import numpy as np
Define the curve fitting function. In your case,
def fit_func(x, a, b, c):
# Curve fitting function
return a * x**3 + b * x**2 + c * x # d=0 is implied
Perform the curve fitting,
# Curve fitting
params = curve_fit(fit_func, x, y)
[a, b, c] = params[0]
x_fit = np.linspace(x[0], x[-1], 100)
y_fit = a * x_fit**3 + b * x_fit**2 + c * x_fit
Plot the results if you please,
plt.plot(x, y, '.r') # Data
plt.plot(x_fit, y_fit, 'k') # Fitted curve
It does not answer the question in the sense that it uses numpy's polyfit function to pass through the origin, but it solves the problem.
Hope someone finds it useful :)
You can use np.linalg.lstsq and construct your coefficient matrix manually. To start, I'll create the example data x and y, and the "exact fit" y0:
import numpy as np
import matplotlib.pyplot as plt
x = np.arange(100)
y0 = 0.07 * x ** 3 + 0.3 * x ** 2 + 1.1 * x
y = y0 + 1000 * np.random.randn(x.shape[0])
Now I'll create a full cubic polynomial 'training' or 'independent variable' matrix that includes the constant d column.
XX = np.vstack((x ** 3, x ** 2, x, np.ones_like(x))).T
Let's see what I get if I compute the fit with this dataset and compare it to polyfit:
p_all = np.linalg.lstsq(X_, y)[0]
pp = np.polyfit(x, y, 3)
print np.isclose(pp, p_all).all()
# Returns True
Where I've used np.isclose because the two algorithms do produce very small differences.
You're probably thinking 'that's nice, but I still haven't answered the question'. From here, forcing the fit to have a zero offset is the same as dropping the np.ones column from the array:
p_no_offset = np.linalg.lstsq(XX[:, :-1], y)[0] # use [0] to just grab the coefs
Ok, let's see what this fit looks like compared to our data:
y_fit = np.dot(p_no_offset, XX[:, :-1].T)
plt.plot(x, y0, 'k-', linewidth=3)
plt.plot(x, y_fit, 'y--', linewidth=2)
plt.plot(x, y, 'r.', ms=5)
This gives this figure,
WARNING: When using this method on data that does not actually pass through (x,y)=(0,0) you will bias your estimates of your output solution coefficients (p) because lstsq will be trying to compensate for that fact that there is an offset in your data. Sort of a 'square peg round hole' problem.
Furthermore, you could also fit your data to a cubic only by doing:
p_ = np.linalg.lstsq(X_[:1, :], y)[0]
Here again the warning above applies. If your data contains quadratic, linear or constant terms the estimate of the cubic coefficient will be biased. There can be times when - for numerical algorithms - this sort of thing is useful, but for statistical purposes my understanding is that it is important to include all of the lower terms. If tests turn out to show that the lower terms are not statistically different from zero that's fine, but for safety's sake you should probably leave them in when you estimate your cubic.
Best of luck!
So I've got some data stored as two lists, and plotted them using
plot(datasetx, datasety)
Then I set a trendline
trend = polyfit(datasetx, datasety)
trendx = []
trendy = []
for a in range(datasetx[0], (datasetx[-1]+1)):
trendx.append(a)
trendy.append(trend[0]*a**2 + trend[1]*a + trend[2])
plot(trendx, trendy)
But I have a third list of data, which is the error in the original datasety. I'm fine with plotting the errorbars, but what I don't know is using this, how to find the error in the coefficients of the polynomial trendline.
So say my trendline came out to be 5x^2 + 3x + 4 = y, there needs to be some sort of error on the 5, 3 and 4 values.
Is there a tool using NumPy that will calculate this for me?
I think you can use the function curve_fit of scipy.optimize (documentation). A basic example of the usage:
import numpy as np
from scipy.optimize import curve_fit
def func(x, a, b, c):
return a*x**2 + b*x + c
x = np.linspace(0,4,50)
y = func(x, 5, 3, 4)
yn = y + 0.2*np.random.normal(size=len(x))
popt, pcov = curve_fit(func, x, yn)
Following the documentation, pcov gives:
The estimated covariance of popt. The diagonals provide the variance
of the parameter estimate.
So in this way you can calculate an error estimate on the coefficients. To have the standard deviation you can take the square root of the variance.
Now you have an error on the coefficients, but it is only based on the deviation between the ydata and the fit. In case you also want to account for an error on the ydata itself, the curve_fit function provides the sigma argument:
sigma : None or N-length sequence
If not None, it represents the standard-deviation of ydata. This
vector, if given, will be used as weights in the least-squares
problem.
A complete example:
import numpy as np
from scipy.optimize import curve_fit
def func(x, a, b, c):
return a*x**2 + b*x + c
x = np.linspace(0,4,20)
y = func(x, 5, 3, 4)
# generate noisy ydata
yn = y + 0.2 * y * np.random.normal(size=len(x))
# generate error on ydata
y_sigma = 0.2 * y * np.random.normal(size=len(x))
popt, pcov = curve_fit(func, x, yn, sigma = y_sigma)
# plot
import matplotlib.pyplot as plt
fig = plt.figure()
ax = fig.add_subplot(111)
ax.errorbar(x, yn, yerr = y_sigma, fmt = 'o')
ax.plot(x, np.polyval(popt, x), '-')
ax.text(0.5, 100, r"a = {0:.3f} +/- {1:.3f}".format(popt[0], pcov[0,0]**0.5))
ax.text(0.5, 90, r"b = {0:.3f} +/- {1:.3f}".format(popt[1], pcov[1,1]**0.5))
ax.text(0.5, 80, r"c = {0:.3f} +/- {1:.3f}".format(popt[2], pcov[2,2]**0.5))
ax.grid()
plt.show()
Then something else, about using numpy arrays. One of the main advantages of using numpy is that you can avoid for loops because operations on arrays apply elementwise. So the for-loop in your example can also be done as following:
trendx = arange(datasetx[0], (datasetx[-1]+1))
trendy = trend[0]*trendx**2 + trend[1]*trendx + trend[2]
Where I use arange instead of range as it returns a numpy array instead of a list.
In this case you can also use the numpy function polyval:
trendy = polyval(trend, trendx)
I have not been able to find any way of getting the errors in the coefficients that is built in to numpy or python. I have a simple tool that I wrote based on Section 8.5 and 8.6 of John Taylor's An Introduction to Error Analysis. Maybe this will be sufficient for your task (note the default return is the variance, not the standard deviation). You can get large errors (as in the provided example) because of significant covariance.
def leastSquares(xMat, yMat):
'''
Purpose
-------
Perform least squares using the procedure outlined in 8.5 and 8.6 of Taylor, solving
matrix equation X a = Y
Examples
--------
>>> from scipy import matrix
>>> xMat = matrix([[ 1, 5, 25],
[ 1, 7, 49],
[ 1, 9, 81],
[ 1, 11, 121]])
>>> # matrix has rows of format [constant, x, x^2]
>>> yMat = matrix([[142],
[168],
[211],
[251]])
>>> a, varCoef, yRes = leastSquares(xMat, yMat)
>>> # a is a column matrix, holding the three coefficients a, b, c, corresponding to
>>> # the equation a + b*x + c*x^2
Returns
-------
a: matrix
best fit coefficients
varCoef: matrix
variance of derived coefficents
yRes: matrix
y-residuals of fit
'''
xMatSize = xMat.shape
numMeas = xMatSize[0]
numVars = xMatSize[1]
xxMat = xMat.T * xMat
xyMat = xMat.T * yMat
xxMatI = xxMat.I
aMat = xxMatI * xyMat
yAvgMat = xMat * aMat
yRes = yMat - yAvgMat
var = (yRes.T * yRes) / (numMeas - numVars)
varCoef = xxMatI.diagonal() * var[0, 0]
return aMat, varCoef, yRes