python - Normalize slices of a ndarray -

i have 3 columns array. first column of array have values between 1 , 10. need extract lines first column 1 , normalize third column of slice of array. repeat same thing rows first column equal 2 etc.

if run code, leaves array unchanged:

for u in np.unique(x[:,0]):     mask= x[:, 0] == u     x[mask][:,2]=x[mask][:,2]/np.sum((x[mask][:,2])) 

if run other slice of code, see r (i placed print r in loop) work want. point original array x unchanged.

for u in np.unique(x[:,0]):     r = x[x[:, 0] == u]     r[:,2]=r[:,2]/np.sum((x[x[:,0]==u][:,2])) 

why that? doing wrong???

here's alternative vectorized approach performance in mind solve problem using np.unique , np.bincount -

tags = np.unique(x[:,0], return_inverse=1)[1] x[:,2] /= np.bincount(tags, x[:,2])[tags] 

to further boost performance, 1 can avoid use of np.unique , directly compute equivalent of np.bincount(tags, xc[:,2]), while making use of fact numbers in first column between 1 , 10, -

np.bincount(xc[:,0].astype(int), xc[:,2], minlength=11)[1:] 

to replace tags, use first column, -

tags = xc[:,0].astype(int)-1