Closed brenmous closed 4 years ago
Training and cross validation are parallelised but prediction doesn't appear to be. Will look at implementing and testing in future.
Prediction is run in parallel - image data is split across nodes. Fixed a bug where random seed was tied to processor rank. This meant that MRF results would vary based on the number of processors being used.
Need to investigate how
multirandomforest
works in parallel. It's currently unknown whether or not it's parallelised.