Closed Jiangshuyi0V0 closed 8 months ago
In the multi_pipeline.sh, it seems like multiple seeds will be chosen in the training process. Is there any motivation behind this setting?
Hey, this is meant to compute statistics like mean/std over a certain number of runs for a given scene, hence we used different seeds for that
In the multi_pipeline.sh, it seems like multiple seeds will be chosen in the training process. Is there any motivation behind this setting?