How to get max/min of "ndarray of dtype[void]"

I have <class ‘numpy.ndarray’> of many <class ‘numpy.dtype[void]’> … It’s from mql5 integration and they are include int/float data, example of single row:

(1649854500, 0.83254, 0.83264, 0.83231, 0.83256, 290, 18, 0)
(1649819400, 1.64308, 1.64314, 1.64259, 1.64262, 185, 57, 0)
… e.t.c.

But I was get many problem with serching max/min values in the axis=1 and many errors with datatype… How to get max value of second column in the specific range?

Full len of ndarray is 100+, but I need get max from second column between 25 - 50 indexes (rows, for example, really it’s various indexes each time) …

And I don’t know maybe need to change type as first step (for dict/list), maybe it’s possible with this data … plz help.

p.s. I was use it, numpy.ndarray.max — NumPy v1.22 Manual
but don’t know how to change type for correct for “void” rows =( And how to set conditions too, but may be it will be more easy if it will be work

What did you use to get the max?

What error did you get?

a = StochasticData.max(axis=1)

TypeError: cannot perform reduce with flexible type

a = StochasticData.astype(np.float).max(axis=1)

TypeError: Cannot cast array data from dtype ([(‘time’, ‘<i8’), (‘open’, ‘<f8’), (‘high’, ‘<f8’), (‘low’, ‘<f8’), (‘close’, ‘<f8’), (‘tick_volume’, ‘<u8’), (‘spread’, ‘<i4’), (‘real_volume’, ‘<u8’)]) to dtype(‘float64’) according to the rule ‘unsafe’

Maybe need to do it with another method, but I don’t know how, and not see special attributes for change type only for column …