search

autonomousvision / convolutional_occupancy_networks

[ECCV'20] Convolutional Occupancy Networks
https://pengsongyou.github.io/conv_onet
MIT License
835 stars 114 forks source link

Config Files for Training From Scratch #6

Closed SrinjaySarkar closed 4 years ago

SrinjaySarkar commented 4 years ago

Are you planning to release the config files for training the models from scratch on different datasets? I am trying to reproduce the results for Shapenet in 32**3 grid but the config file for training is not present in the repo and the generation config files does not contain all the fields as the training file.

pengsongyou commented 4 years ago

Hi, the training config for 32**3 grid is in the repo: https://github.com/autonomousvision/convolutional_occupancy_networks/blob/master/configs/pointcloud/shapenet_grid32.yaml

SrinjaySarkar commented 4 years ago

Sorry, if this is a dumb question , I am having trouble writing the get_dataset and get_generator functions. Both functions for conv_onet contain lines like train': cfg['data']['train_split'] and sample=cfg['generation']['use_sampling'] and these subfields don't exist in the YAML file for Shapenent_grid32. There are a couple of other field like this in both the functions. Do I fill in a default value for this or is there any other file I am missing?

SrinjaySarkar commented 4 years ago

Just read the update_recursive carefully. Thank you for your prompt reply!!