Closed ResonWang closed 1 year ago
self.model.apply(init_yolo) has no effect on the vid model for we will load pretrain-weights. In our experiment, fix_bn could make training more stable. But we find its performance varies for different detectors and datasets.
Dear author, I noticed the following code:
def fix_bn(m):
In your experiments, is there big difference between self.model.apply(init_yolo) and self.model.apply(fix_bn) ?