Hi all,
I am running 3D classification and I am attempting to use the outputs of alignments3D/error, for 3d classification with 2 classes.
According to this forum post, alignments3D/error should have the log likelihood values at the optimal pose.
However, the outputs I am getting have all zeros here for both classes. I am also getting zeros for alignments3D/error_min, resid_pow,slice_pow,imag_pow,cross_corr, although I don’t know if its relevant.
Here is a snapshot of the output:
('alignments3D_multi/error', [[0. 0.] [0. 0.] [0. 0.] ... [0. 0.] [0. 0.] [0. 0.]]),
('alignments3D_multi/error_min', [[0. 0.] [0. 0.] [0. 0.] ... [0. 0.] [0. 0.] [0. 0.]]),
('alignments3D_multi/resid_pow', [[0. 0.] [0. 0.] [0. 0.] ... [0. 0.] [0. 0.] [0. 0.]]),
('alignments3D_multi/slice_pow', [[0. 0.] [0. 0.] [0. 0.] ... [0. 0.] [0. 0.] [0. 0.]]),
('alignments3D_multi/image_pow', [[0. 0.] [0. 0.] [0. 0.] ... [0. 0.] [0. 0.] [0. 0.]]),
('alignments3D_multi/cross_cor', [[0. 0.] [0. 0.] [0. 0.] ... [0. 0.] [0. 0.] [0. 0.]]),
('alignments3D_multi/alpha', [[1. 1.] [1. 1.] [1. 1.] ... [1. 1.] [1. 1.] [1. 1.]]),
('alignments3D_multi/alpha_min', [[0. 0.] [0. 0.] [0. 0.] ... [0. 0.] [0. 0.] [0. 0.]]),
('alignments3D_multi/weight', [[0. 0.] [0. 0.] [0. 0.] ... [0. 0.] [0. 0.] [0. 0.]]),
('alignments3D_multi/pose_ess', [[0. 0.] [0. 0.] [0. 0.] ... [0. 0.] [0. 0.] [0. 0.]]),
('alignments3D_multi/shift_ess', [[0. 0.] [0. 0.] [0. 0.] ... [0. 0.] [0. 0.] [0. 0.]]),
('alignments3D_multi/class_posterior', [[1. 0.] [0. 1.] [1. 0.] ... [0. 1.] [1. 0.] [0. 1.]]),
('alignments3D_multi/class', [[0 1] [0 1] [0 1] ... [0 1] [0 1] [0 1]]),
('alignments3D_multi/class_ess', [1. 1. 1. ... 1. 1. 1.]),
Should I be getting all zeros here for error? I am wondering if there is an output error (maybe a user error on my part, I’m very new at this).The classification has run successfully and particles have been assigned, so not sure what the problem is.
Thanks in advance, first time user here so might likely just be an issue on my part!
cryosparc version: 4.6