CodeFormer pre-trained VQGAN of 256*256 resolution ratio

It takes too long to train STAGE1(about 16days), so I want to add my own dataset for fine-tuning. Can you release pre-trained VQGAN model in STAGE1? Thanks a lot~

Nov 21 '23 04:11 create-li

It takes too long to train STAGE1(about 16days), so I want to add my own dataset for fine-tuning. Can you release pre-trained VQGAN model in STAGE1? Thanks a lot~

hi,I want to train 256x256 also.But when the fisrt stage，I get erros as following： Traceback (most recent call last): File "basicsr/train.py", line 225, in train_pipeline(root_path) File "basicsr/train.py", line 186, in train_pipeline model.optimize_parameters(current_iter) File "/mnt/8T_disk/z/data/extra/virtual_person/CodeFormer/basicsr/models/vqgan_model.py", line 153, in optimize_parameters last_layer = self.net_g.module.generator.blocks[-1].weight File "/home/z/data/miniconda3/envs/codeformer/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1614, in getattr raise AttributeError("'{}' object has no attribute '{}'".format( AttributeError: 'VQAutoEncoder' object has no attribute 'module' Could you give me some hints？

Jan 06 '24 03:01 franklydog

I have encountered it, just modify the number of model blocks

Feb 27 '24 06:02 create-li

I have encountered it too .How to modify the number of model blocks?

Aug 28 '24 01:08 volcanoxs