Possible unbalanced tuple unpacking with sequence in scipy optimization curve_fit

Question

I got error from pylint from my code. I don’t know how to fix that. can you please help me?

The code is here:

import numpy as np
import matplotlib.pyplot as plt
from scipy.optimize import curve_fit

# Define the two points
x1, y1 = 4, 0
x2, y2 = 15, 1

# Define the square root function
def sqrt_func(x, a, b):
    return a * np.sqrt(x - x1) + y1

# Fit the curve to the two points
popt, pcov = curve_fit(sqrt_func, [x1, x2], [y1, y2])

# Generate intermediate x values between 4 and 15
x_values = np.linspace(4, 15, num=100)

# Use the fitted curve to calculate y values
y_values = sqrt_func(x_values, *popt)

# Plot the curve and the two points
plt.plot(x_values, y_values)
plt.scatter([x1, x2], [y1, y2])
plt.show()

in this bellow line I have this error: ** Possible unbalanced tuple unpacking with sequence defined at line 885 of scipy.optimize._minpack_py: left side has 2 labels, right side has 5 values **

popt, pcov = curve_fit(sqrt_func, [x1, x2], [y1, y2])

Asked By: Mohammadreza Shahsavar

||

Source

Answer 1

because the curve_fit function returns a tuple of two elements, but you are trying to unpack it into two variables popt and pcov.

You can modify the line to just unpack the first element of the tuple, which is popt, and ignore the second element pcov by using the underscore _ as a placeholder variable:

popt, _ = curve_fit(sqrt_func, [x1, x2], [y1, y2])

Or

# pylint: disable=unpacking-non-sequence
popt, pcov = curve_fit(sqrt_func, [x1, x2], [y1, y2])

This tells pylint to ignore the "Possible unbalanced tuple unpacking" error

Answered By: ma9

Answer 2

Here’s the relevant source code part:


def curve_fit(f, xdata, ydata, p0=None, sigma=None, absolute_sigma=False,
              check_finite=None, bounds=(-np.inf, np.inf), method=None,
              jac=None, *, full_output=False, nan_policy=None,
              **kwargs):
    ... # a lot of code here
    
    if full_output:
        return popt, pcov, infodict, errmsg, ier
    else:
        return popt, pcov

pylint analyses the body and understands that curve_fit can return a 2-tuple or 5-tuple, but fails to infer the relationship with full_output input parameter. We’re more capable then pylint and can read the definition to find out that the return type is always 2-tuple in your case, and so pylint gave a false positive. You can add a comment to explain the ignorance reason and to suppress the error message, like this:

# with full_output = False, always returns a 2-tuple
# pylint: disable-next=unbalanced-tuple-unpacking
popt, pcov = curve_fit(sqrt_func, [x1, x2], [y1, y2])

Note that I’m using pylint: disable-next, see this question for reasons to prefer it to pylint: ignore in many cases.

Answered By: SUTerliakov

Possible unbalanced tuple unpacking with sequence in scipy optimization curve_fit

Question:

Answers: