Generators in python

Question

I just came across the problem that I could not use the built-in range() function of python for float values. So I decided to use a float-range function that I manually defined:

def float_range(start, stop, step):
    r = start
    while r < stop:
        yield r
        r += step

Problem is: This function returns a generator object. Now I needed this range for a plot:

ax.hist(some_vals, bins=(float_range(some_start, some_stop, some_step)), normed=False)

This will not work due to the return type of float_range(), which is a generator, and this is not accepted by the hist() function. Is there any workaround for me except not using yield?

Why is not using yield a "workaround"? What did you hope to gain by using yield? — BrenBarn
– BrenBarn, Commented Jun 2, 2014 at 7:34
yield is the standard solution for defining a default range function. — PKlumpp
– PKlumpp, Commented Jun 2, 2014 at 7:35

Ashalynd · Accepted Answer · 2014-06-02 07:36:48Z

3

If you need to give a range, you can do either

ax.hist(some_vals, bins=(list(float_range(some_start, some_stop, some_step))), normed=False)

Or just make your float_range return a list:

def float_range(start, stop, step):
    result = []
    r = start
    while r < stop:
        result.append(r)
        r += step
    return result

BTW: nice reference on what you can do with generators in Python: https://wiki.python.org/moin/Generators

answered Jun 2, 2014 at 7:36

Ashalynd

12.6k2 gold badges36 silver badges38 bronze badges

Sign up to request clarification or add additional context in comments.

1 Comment

PKlumpp Over a year ago

that list()-thing is what I was looking for, thank you!

Davidmh · Accepted Answer · 2014-06-02 07:42:56Z

1

You can convert a generator to a list using:

list(float_range(1, 4, 0.1))

But seeing you are using matplotlib, you should be using numpy's solution instead:

np.arange(1, 4, 1, dtype=np.float64)

or:

np.linspace(start, stop, (start-stop)/step)

edited Jun 2, 2014 at 7:42

answered Jun 2, 2014 at 7:37

Davidmh

3,87521 silver badges38 bronze badges

4 Comments

PKlumpp Over a year ago

but with np.arange I could not get any float values to work, like np.arange(0.2, 0.8, 0.05)

user2357112 Over a year ago

NumPy's solution is linspace. It is specifically warned that using arange for non-integer steps is a bad idea in the documentation.

Davidmh Over a year ago

@ZerO you can force it to give you floats.

Davidmh Over a year ago

@user2357112 using arange with floats may lead to numerical inaccuracies, and so not give you exactly what you want; but the OP will have the same problem with the proposed approach. Using one or the other depends on what do you want to ensure are correct: distance or number of elements. For the bins of a plot, probably either would work fine.

Collectives™ on Stack Overflow

Generators in python

2 Answers 2

1 Comment

4 Comments

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

1 Comment

4 Comments

Related