xulabs

i get similar error. i guess it is due to the last minibatch has an odd number of samples, however the paralleled model only produced even number of predictions

@YuchenZeng Please take a look at itk-vtk-viewer, and discuss with @ijinjay

@YuchenZeng add the three.js version under aitom/pick/plot/particle_location_display__three_js

Example code can be found at aitom/tomominer/average/genetic_algorithm/aligned_refine aitom/tomominer/pursuit/multi/util.vol_avg__local https://github.com/alberlab/tomominer/blob/master/bin/tm_average To fully utilize all cpu cores, need to be parallelized using multiprocessing A wrapper function can be found here: aitom.parallel.multiprocessing.util

Alignment based classification method, see following paper Xu M, Beck M, Alber F. High-throughput subtomogram alignment and classification by Fourier space constrained fast volumetric matching. Journal of Structural Biology. 2012...

@evaxguo thanks for the suggestion. You are welcome to discuss with other team members to help to implement this feature.

@geoffwoollard generally only the fast subtomogram alignment part involves c++ code. If you don't use the subtomogram alignment, then you can just clone the library without compiling it. The c++...

Feature decomposition based saliency detection.   

Thanks @xavierrollet for the suggestion.

Thanks @xavierrollet .