DOC: Improve docstring of Series/DataFrame.bool

[datapythonista]

Not a big fan of this method, I'd deprecate it, in favor of bool(df.squeeze()). But since it's quite tricky, let's at least have the docstring as clear as possible.

[jorisvandenbossche]

Small comment (and I agree that this method doesn't sound very useful (never used it myself), there is also .item() for Series that does exactly the same but for all data types and not just bool)

[jorisvandenbossche]

Suggested change

	numpy.bool : NumPy boolean data type, used by pandaas for boolean values.
	numpy.bool : NumPy boolean data type, used by pandas for boolean values.

Two notes: this is not actually the boolean data type (but an alias for built-in bool, the numpy one is np.bool_), and is this reference actually useful? (the method returns a builtin bool, not a numpy one)

[datapythonista]

Good points, thanks for the feedback.

What I had in mind is that given the name of the method, users may come to this page looking for a way to convert a Series or a DataFrame to boolean (like elementwise). I was trying to add references that may be useful for them. Didn't think of item, but may be could be added here too. Also, not sure if any and all could be useful to some visitors.

Also happy to remove 'numpy.bool_`, it's difficult to draw the line of what could be useful and what is too much.

What would you have?

[jorisvandenbossche]

Sorry for the slow reply, fixing it up to refer to np.bool_ instead of np.bool is fine for me

[jreback]

can you merge master and update

[jorisvandenbossche]

Thanks!