detect if variable is of sympy type

问题

I have a variable which may or may not be a sympy class. I want to convert it to a float but I’m having trouble doing this in a general manner:

$ python
Python 2.7.3 (default, Dec 18 2014, 19:10:20) 
[GCC 4.6.3] on linux2
>>> import sympy
>>> x = sympy.sqrt(2)
>>> type(x)
<class 'sympy.core.power.Pow'>
>>> y = 1 + x
>>> type(y)
<class 'sympy.core.add.Add'>
>>> z = 3 * x
>>> type(z)
<class 'sympy.core.mul.Mul'>
>>> if isinstance(z, sympy.core):
...     z = z.evalf(50) # 50 dp
... 
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
TypeError: isinstance() arg 2 must be a class, type, or tuple of classes and types

I will be testing x, y, z to convert into a float. note that I can't just run evalf() on x, y and z without checking because they might be integers or floats already and this would raise an exception.

---

sympy.sympify() unfortunately does not convert to float. if it did then that would be the ideal solution to my problem:

>>> sympy.sympify(z)
3*sqrt(2)
>>> type(sympy.sympify(z))
<class 'sympy.core.mul.Mul'>

回答1:

All sympy objects inherit from sympy.Basic. To evaluate an sympy expression numerically, just use .n()

In [1]: import sympy

In [2]: x = sympy.sqrt(2)

In [3]: x
Out[3]: sqrt(2)

In [4]: x.n()
Out[4]: 1.41421356237310

In [5]: if (isinstance(x, sympy.Basic)):
   ...:     print(x.n())
   ...:     
1.41421356237310

回答2:

heh, at the end of the day, it turns out you can just convert a sympy object to float using python's native float conversion function! don't know why i didn't try this straight away...

>>> import sympy
>>> x = sympy.sqrt(2)
>>> float(x)
1.4142135623730951

i will use this since it works even if x is a python int or already a python float

来源：https://stackoverflow.com/questions/30023064/detect-if-variable-is-of-sympy-type