We Keep Coding

why am I getting OptimizeError while trying to fit a gaussian/lorentzian to data using curve_fit?

I am trying to fit a data set which may fit a gaussian or lorentzian, with scipy optimize curve_fit function.
I am getting the error:
"OptimizeWarning: Covariance of the parameters could not be estimated
warnings.warn('Covariance of the parameters could not be estimated',"
the data set looks like this:
enter image description here
which , as you can see, may fit a gaussian.
my code is :
def gaussian(x,a,b,c,d):
func=a*np.exp(-((x-b)**2)/c)+d
return func
def lorentzian (x,a,b,c):
func=a/(((x-b)**2+a**2)*np.pi)+c
return func
x,y_data= np. loadtxt('in 0.6 out 0.6.dat', unpack = True)
popt, pcov = curve_fit(lorentzian, x, y_data)
thank you!

You're getting this error because the fitting algorithm couldn't find an appropriate solution. No matter where it moved the parameters, the fit quality didn't change. If you provide an initial guess, you're more likely to reach the solution. Given that the function parameters are relatively easily obtained from glancing the curves, you could provide most of them. For example, the center (which you called a) is around 545.5. Wikipedia also has a relationship for the value at the maximum for a slightly different form of your equation, which lacks the c parameter to shift the curve upwards. Providing the guess p0 = (0.1, 545.5, 0) and a bound of (0, 1E10) you get something much closer to your results, yet still unsatisfactory (next time, provide the data array, I had to use a point extractor to plot this)
Now, notice how you're supposed to reach a maximum value of 40, yet that seems unattainable by your model. I took the liberty of normalizing your model simply by dividing it by its maximum value and trying to fit again. I don't know if this is the appropriate normalization, but this is just to illustrate the difference. This time, the result is much more satisfactory:
Yet I think a lorentzian is a bit too narrow for your curve (especially evident if you set c to 0), which looks much more like a gaussian (given you provided its definition but didn't use it, I guess you probably would have used it in the future).
Note how I didn't have to normalize y.
In summary:
Provide an initial guess to your fitting algorithm, and bounds if possible.
Always plot your models and data to see what's going on.
Be aware of the limits of your models, what values it can or can't reach. Use this to evaluate if a fit is even possible in the first place.

Why the initial value p0 you assigned will largely affect the fitting result

When I assign different initial values to P0, the fitting varies two orders of magnitude. The question is, how could I find the right one as the purpose of the fitting is the search the right parameters.
I got 2 variables and two unknown parameters to be found by a suitable method.
p0 = [0.5,0.3] #initial value
c,cov = curve_fit(titrition,(xp,xd),y,p0)
yp = titrition ((xp,xd),c[0],c[1])

This question isn't so much a coding question, but a very specific numerics question. curve_fit is simply trying to minimize a cost function to fit your function to the data the best it can. I don't know what the function titration looks like, but I suspect that optimizing it is non-convex. In other words, curve_fit is probably finding a different optimum value depending on the initial guess you supply. This is where you must use some sort of physical intuition to supply a best guess for what the initial conditions should be.

how to let the fit function to define the fitting range?

I want to fit part of my data with python, using lmfit( it is not a must!). I'd like to have dynamic range of data to be fitted, meaning having two fitting parameters which define the part of my data to be fitted (let's call it lower and upper boundaries).
the reason is I have many data sets. in each of the the fitting range varies and I cannot define a model to fit the whole range of data. on the other hand I cannot go through each data set and define the fitting range.
is it possible at all? I thought of multiplying a pulse function to my model which affects the original data as well. though as far as I understand I cannot tell lmfit to multiply it to the data. so I am out of idea!

The number of observations (data points or length of the residual array) returned by the model function or function to be minimized has to be the same throughout an individual fit. Of course, this can change between successive fits. So, you could try multiple fits for each data set with different ranges, perhaps set based on the previous fit.
I think that your idea of using "where the fit is bad" to determine where to not fit is somewhat suspect and you would want to make sure to avoid that leading to absurd results. If for example, the range was automatically reduced so much so that Ndata = Nvariables+1, you could probably get a very low chi-square compared to Ndata = 100*Nvariables.
Without knowing the particulars, I think you would better off coming up with criteria for selecting the data range that depended on the data alone, and not a fit to it.

How to create dataset for fitting a function in scipy stats?

I want to fit some data to a Pareto distribution using the scipy.stats library. I am not sure if the issue might be numerical, so just to be safe; I have values measured for the dependent variable (let's call them 'pushes') for the independent variable ('minutes') starting at a few thousand minutes and every ten minutes thereafter (with the exception of a few points that were removed during data cleaning).
e.g.
2780.0 362.0
2800.0 376.0
2810.0 393.0
...
The best info I can find says something like
from scipy.stats import pareto
result = pareto.fit(data)
and I have no idea how this data is to be formatted in this case. I've tried the following but all result in errors.
result = pareto.fit(zip(minutes, pushes))
result = pareto.fit(pushes)
The error is usually
Warning: invalid value encountered in double_scalars
would greatly appreciate some guidance, thank you.

As I mentioned in the comments above, pareto.fit() is not what you're looking for.
The .fit() methods of the continuous distributions in scipy.stats obtain an estimate of the parameters of the distribution that maximise the probability of observing some particular set of sample values. Therefore, pareto.fit() wants only a single array argument containing the samples you want to fit the distribution to. The other keyword arguments control various aspects of the fitting process, for example by specifying initial values for the distribution parameters.
What you're actually trying to do is to fit the relationship between some independent variable x and some dependent variable y, i.e.
y_fit = f(x, params)
What you need to do is:
Choose some functional form for f. From your description, the plot of y vs x resembles the probability density function for a Pareto distribution, so perhaps either this or a decaying exponential might be appropriate.
Find the set of params that minimize some measure of the difference between y and y_fit (usually the sum of squared differences). You could use scipy.optimize.curve_fit or scipy.optimize.minimize to do this.

Maximum Likelihood Estimate pseudocode

I need to code a Maximum Likelihood Estimator to estimate the mean and variance of some toy data. I have a vector with 100 samples, created with numpy.random.randn(100). The data should have zero mean and unit variance Gaussian distribution.
I checked Wikipedia and some extra sources, but I am a little bit confused since I don't have a statistics background.
Is there any pseudo code for a maximum likelihood estimator? I get the intuition of MLE but I cannot figure out where to start coding.
Wiki says taking argmax of log-likelihood. What I understand is: I need to calculate log-likelihood by using different parameters and then I'll take the parameters which gave the maximum probability. What I don't get is: where will I find the parameters in the first place? If I randomly try different mean & variance to get a high probability, when should I stop trying?

I just came across this, and I know its old, but I'm hoping that someone else benefits from this. Although the previous comments gave pretty good descriptions of what ML optimization is, no one gave pseudo-code to implement it. Python has a minimizer in Scipy that will do this. Here's pseudo code for a linear regression.
# import the packages
import numpy as np
from scipy.optimize import minimize
import scipy.stats as stats
import time
# Set up your x values
x = np.linspace(0, 100, num=100)
# Set up your observed y values with a known slope (2.4), intercept (5), and sd (4)
yObs = 5 + 2.4*x + np.random.normal(0, 4, 100)
# Define the likelihood function where params is a list of initial parameter estimates
def regressLL(params):
# Resave the initial parameter guesses
b0 = params[0]
b1 = params[1]
sd = params[2]
# Calculate the predicted values from the initial parameter guesses
yPred = b0 + b1*x
# Calculate the negative log-likelihood as the negative sum of the log of a normal
# PDF where the observed values are normally distributed around the mean (yPred)
# with a standard deviation of sd
logLik = -np.sum( stats.norm.logpdf(yObs, loc=yPred, scale=sd) )
# Tell the function to return the NLL (this is what will be minimized)
return(logLik)
# Make a list of initial parameter guesses (b0, b1, sd)
initParams = [1, 1, 1]
# Run the minimizer
results = minimize(regressLL, initParams, method='nelder-mead')
# Print the results. They should be really close to your actual values
print results.x
This works great for me. Granted, this is just the basics. It doesn't profile or give CIs on the parameter estimates, but its a start. You can also use ML techniques to find estimates for, say, ODEs and other models, as I describe here.
I know this question was old, hopefully you've figured it out since then, but hopefully someone else will benefit.

If you do maximum likelihood calculations, the first step you need to take is the following: Assume a distribution that depends on some parameters. Since you generate your data (you even know your parameters), you "tell" your program to assume Gaussian distribution. However, you don't tell your program your parameters (0 and 1), but you leave them unknown a priori and compute them afterwards.
Now, you have your sample vector (let's call it x, its elements are x[0] to x[100]) and you have to process it. To do so, you have to compute the following (f denotes the probability density function of the Gaussian distribution):
f(x[0]) * ... * f(x[100])
As you can see in my given link, f employs two parameters (the greek letters µ and σ). You now have to calculate the values for µ and σ in a way such that f(x[0]) * ... * f(x[100]) takes the maximum possible value.
When you've done that, µ is your maximum likelihood value for the mean, and σ is the maximum likelihood value for standard deviation.
Note that I don't explicitly tell you how to compute the values for µ and σ, since this is a quite mathematical procedure I don't have at hand (and probably I would not understand it); I just tell you the technique to get the values, which can be applied to any other distributions as well.
Since you want to maximize the original term, you can "simply" maximize the logarithm of the original term - this saves you from dealing with all these products, and transforms the original term into a sum with some summands.
If you really want to calculate it, you can do some simplifications that lead to the following term (hope I didn't mess up anything):
Now, you have to find values for µ and σ such that the above beast is maximal. Doing that is a very nontrivial task called nonlinear optimization.
One simplification you could try is the following: Fix one parameter and try to calculate the other. This saves you from dealing with two variables at the same time.

You need a numerical optimisation procedure. Not sure if anything is implemented in Python, but if it is then it'll be in numpy or scipy and friends.
Look for things like 'the Nelder-Mead algorithm', or 'BFGS'. If all else fails, use Rpy and call the R function 'optim()'.
These functions work by searching the function space and trying to work out where the maximum is. Imagine trying to find the top of a hill in fog. You might just try always heading up the steepest way. Or you could send some friends off with radios and GPS units and do a bit of surveying. Either method could lead you to a false summit, so you often need to do this a few times, starting from different points. Otherwise you may think the south summit is the highest when there's a massive north summit overshadowing it.

As joran said, the maximum likelihood estimates for the normal distribution can be calculated analytically. The answers are found by finding the partial derivatives of the log-likelihood function with respect to the parameters, setting each to zero, and then solving both equations simultaneously.
In the case of the normal distribution you would derive the log-likelihood with respect to the mean (mu) and then deriving with respect to the variance (sigma^2) to get two equations both equal to zero. After solving the equations for mu and sigma^2, you'll get the sample mean and sample variance as your answers.
See the wikipedia page for more details.