search

VITA-Group / EnlightenGAN

[IEEE TIP] "EnlightenGAN: Deep Light Enhancement without Paired Supervision" by Yifan Jiang, Xinyu Gong, Ding Liu, Yu Cheng, Chen Fang, Xiaohui Shen, Jianchao Yang, Pan Zhou, Zhangyang Wang
Other
918 stars 200 forks source link

Training problems #1

Closed TerryYiDa closed 5 years ago

TerryYiDa commented 5 years ago

Thank you for your excellent work， I have some questions:

  1. Is the pre-training model open source?
  2. model block order is CONV->leakyrelu-> bn, why not is conv-bn-leakyrelu?
laomaup commented 5 years ago

thank you for sharing, i have a problem. ModuleNotFoundError: No module named 'psnr' when running test.py.

yifanjiang19 commented 5 years ago

I will open source other guidelines soon, please wait a few days.

yifanjiang19 commented 5 years ago

@TerryYiDa @laomaup Hi, I have uploaded the specific command. Please contact me if you have other questions.

UtopiaHu commented 5 years ago

@yueruchen Thank you for your work. I couldn't find the requirements.txt file. By the way, where is the supplementary materials mentioned in your paper ?

yifanjiang19 commented 5 years ago

@UtopiaHu Sorry, I add it now.

laomaup commented 5 years ago

i am runningTesting process, python scipts/script.py --predict this is the problem: CustomDatasetDataLoader Traceback (most recent call last): File "predict.py", line 16, in data_loader = CreateDataLoader(opt) File "/data/mypycode/EnlightenGAN-master/data/data_loader.py", line 6, in CreateDataLoader data_loader.initialize(opt) File "/data/mypycode/EnlightenGAN-master/data/custom_dataset_data_loader.py", line 39, in initialize self.dataset = CreateDataset(opt) File "/data/mypycode/EnlightenGAN-master/data/custom_dataset_data_loader.py", line 11, in CreateDataset from data.unaligned_dataset import UnalignedDataset File "/data/mypycode/EnlightenGAN-master/data/unaligned_dataset.py", line 4, in import torchvision.transforms as transforms File "/home/xianjin/.conda/envs/allpose/lib/python3.6/site-packages/torchvision/init.py", line 1, in from torchvision import models File "/home/xianjin/.conda/envs/allpose/lib/python3.6/site-packages/torchvision/models/init.py", line 10, in from . import segmentation File "/home/xianjin/.conda/envs/allpose/lib/python3.6/site-packages/torchvision/models/segmentation/init.py", line 1, in from .segmentation import * File "/home/xianjin/.conda/envs/allpose/lib/python3.6/site-packages/torchvision/models/segmentation/segmentation.py", line 1, in from .._utils import IntermediateLayerGetter File "/home/xianjin/.conda/envs/allpose/lib/python3.6/site-packages/torchvision/models/_utils.py", line 7, in class IntermediateLayerGetter(nn.ModuleDict): AttributeError: module 'torch.nn' has no attribute 'ModuleDict'

Can you provide a more detailed file to tell us how to use it? thank you

laomaup commented 5 years ago

Hi, i have a new problem when i try to train. CustomDatasetDataLoader dataset [UnalignedDataset] was created single Traceback (most recent call last): File "predict.py", line 18, in model = create_model(opt) File "/data/disk2/EnlightenGAN-master/models/models.py", line 36, in create_model model.initialize(opt) File "/data/disk2/EnlightenGAN-master/models/single_model.py", line 72, in initialize self.load_network(self.netG_A, 'G_A', which_epoch) File "/data/disk2/EnlightenGAN-master/models/base_model.py", line 53, in load_network network.load_state_dict(torch.load(save_path)) File "/home/xianjin/anaconda3/envs/enlighten/lib/python3.6/site-packages/torch/serialization.py", line 382, in load f = open(f, 'rb') FileNotFoundError: [Errno 2] No such file or directory: './checkpoints/enlightening/200_net_G_A.pth'

where can i find checkpoints and the files.

yifanjiang19 commented 5 years ago

@laomaup This is the pretrained model, you should first train the model and get it. I will upload the pretrained model soon.

yifanjiang19 commented 5 years ago

@TerryYiDa @laomaup @UtopiaHu I've uploaded the pre-trained model and environment requirement, follow the guideline you should get the expected results. Feel free to contact me if you have other questions.