fubinfb / NTF

This is the official PyTorch implementation of the paper “Neural Transformation Fields for Arbitrary-Styled Font Generation”.

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

So many bugs

jackwolfey opened this issue · comments

1、NTFmodel/dataset_NeRFMS.py line,27: have two invalid chars that couldn't execute by python interpreter
2、train_NTF.py line,27: there does not have a class named NTFTrainDataset in NTFmodel/dataset_NeRFMS.py
3、NTFmodel/dataset_NeRFMS.py line,30: always set source_path = " ", and always raise error at next line self.source = read_font(source_path)
4、NTFmodel/NTFtrainer.py line,17: save_tensor_as_mat is nerver used, and clovaai/fewshot-font-generation dose not have this function at all, this phenomenon happend every where, import something but not used in the file
5、in train_NTF.py use absolute import but in submodule use relative import, which cause import error

So, did you guys test these code on you computer before upload to github?

Sorry for this issue, we promise to provide a better code before June 10, 2024.