python-scipyHow do I use Python's Scipy library to minimize a scalar?

To use Python's Scipy library to minimize a scalar, you need to first import the minimize function from the scipy.optimize module.

from scipy.optimize import minimize

Then, define the objective function you want to minimize. This should be a function that takes a single argument and returns the value of the scalar you want to minimize.

def objective_function(x):
    return x ** 2

Next, define the bounds of the scalar you want to minimize. This is done using a tuple of two numbers, where the first number is the lower bound and the second number is the upper bound.

bounds = (-5, 5)

Finally, call the minimize function with the objective function and bounds as arguments. The minimize function returns an object containing the optimal value of the scalar and other information.

result = minimize(objective_function, bounds=bounds)
print(result)

      fun: 1.9242640687119285e-16
 hess_inv: <2x2 LbfgsInvHessProduct with dtype=float64>
      jac: array([-3.55271368e-15,  0.00000000e+00])
  message: b'CONVERGENCE: NORM_OF_PROJECTED_GRADIENT_<=_PGTOL'
     nfev: 6
      nit: 2
   status: 0
  success: True
        x: array([-3.55271368e-15, -3.55271368e-15])

The optimal value of the scalar is 1.9242640687119285e-16, which is the fun attribute of the result object.

Code explanation

**

1. from scipy.optimize import minimize - imports the minimize function from the scipy.optimize module.
2. def objective_function(x): - defines the objective function you want to minimize.
3. bounds = (-5, 5) - defines the bounds of the scalar you want to minimize.
4. result = minimize(objective_function, bounds=bounds) - calls the minimize function with the objective function and bounds as arguments.
5. print(result) - prints the result object containing the optimal value of the scalar and other information.

## Helpful links

1. https://docs.scipy.org/doc/scipy/reference/generated/scipy.optimize.minimize.html
2. https://en.wikipedia.org/wiki/Scalar_(mathematics)

Edit this code on GitHub