search

mit-han-lab / spvnas

[ECCV 2020] Searching Efficient 3D Architectures with Sparse Point-Voxel Convolution
http://spvnas.mit.edu/
MIT License
582 stars 109 forks source link

Config to train the baseline model #72

Closed guochengqian closed 3 years ago

guochengqian commented 3 years ago

Hi,

Thanks for open-souring the great work. May I ask for the configurations for training the baseline models like SPVCNN and Minkunet? Since https://github.com/mit-han-lab/spvnas/issues/40 also mentioned that it can not reproduce the results using the default configuration with SPVCNN.

zhijian-liu commented 3 years ago

The results should be able to be reproduced with the default configuration from https://github.com/mit-han-lab/spvnas/blob/master/configs/semantic_kitti/default.yaml#L16-L31.