search

XuyangBai / PPF-FoldNet

PyTorch reimplementation for "PPF-FoldNet: Unsupervised Learning of Rotation Invariant 3D Local Descriptors" https://arxiv.org/abs/1808.10322
99 stars 19 forks source link

converge issue #7

Closed tangbohu closed 4 years ago

tangbohu commented 4 years ago

Thanks for the excellent work! In my training, the loss in the first epoch is about 10000, and after 20 epochs, the loss decreases to ~1000. Should we add more epochs？

XuyangBai commented 4 years ago

@tangbohu Sorry I implemented this repo a long time ago so I didn't remember the value of the loss. But I usually train for 100 epoch if remember correctly.