array = np.zeros(10)
[0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
array = np.ones((1,3,4))
array = np.full((1,3,4), 9)
[[[9 9 9 9]
[9 9 9 9]
[9 9 9 9]]]
array = np.eye(4)
[[1. 0. 0. 0.]
[0. 1. 0. 0.]
[0. 0. 1. 0.]
[0. 0. 0. 1.]]
# do čtverce
np.empty((2,3))
[[ 1.79769313e+308 -1.79769313e+308 2.00000000e+000]
[ 2.22044605e-016 2.22507386e-308 4.94065646e-324]]
# Rychlejší, ale random balast
np.arange(0,10,2) #start, stop, step
[0 2 4 6 8]
np.linspace(0,10,5) # rozdělí na pět částí
[ 0. 2.5 5. 7.5 10. ]
aggregate functions
np.sum(array)
np.mean(array)
np.median(array)
np.std(array) #rozptyl
np.min(array)
np.max(array)
np.argmin(array) #index
np.argmax(array)
np.sum(array, axis=1) #radky, 0 pro sloupce