ValueError: Buffer dtype mismatch, expected 'float' but got 'double'

[rw1993]

Try below code and get ValueError.

import numpy as np
import pydensecrf.densecrf as dcrf
d = dcrf.DenseCRF(80, 2)
U = np.random.random(size=(2, 80))
d.setUnaryEnergy(U)
Traceback (most recent call last):
File "", line 1, in
d.setUnaryEnergy(U)
File "pydensecrf/densecrf.pyx", line 80, in pydensecrf.densecrf.DenseCRF.setUn
aryEnergy (pydensecrf/densecrf.cpp:3370)
ValueError: Buffer dtype mismatch, expected 'float' but got 'double'

[Nandayang]

@rw1993 Hi, i m facing the same problem. Do you solve this ?

[lucasb-eyer]

People had this error before (#49) and it's a user error. I think the error is pretty clear: you should pass a float, but you're passing a double. Fix by doing U = U.astype(np.float32). I'll add a note to the FAQ in the README.