Syen Park

I believe you can make use of this: https://cv-tricks.com/tensorflow-tutorial/save-restore-tensorflow-models-quick-complete-tutorial/. Or you can reference [evaluate.py](https://github.com/charlesq34/pointnet/blob/master/evaluate.py) in this github by tracking `parser.add_argument('--model_path', default='log/model.ckpt', help='model checkpoint file path [default: log/model.ckpt]')`.

@sherineza, you can make use of the pre-trained model and `hierarchical point set feature learning ` architecture to do extract features.

Hi @moondabaojian, can you try `ds_factors = [4, 4]` instead of [4, 1] and get back to me? FYI, I used [pytorch_coma](https://github.com/pixelite1201/pytorch_coma) to reproduce the work. First of all, I...

@moondabaojian, I suppose some format or header information of `.npz` leads to the slight different size that you point out. For instance, the discrepancy in the format such as `Compressed...