search

ChiWeiHsiao / DeepVO-pytorch

PyTorch Implementation of DeepVO
358 stars 106 forks source link

Hello,The following website cannot be opened. #14

Open xuwei1245645564 opened 5 years ago

xuwei1245645564 commented 5 years ago

Hello,The following Hello,The following website cannot be opened. "https://drive.google.com/file/d/1l0s3rYWgN8bL0Fyofee8IhN-0knxJF22/view" cannot be opened. "https://drive.google.com/file/d/1l0s3rYWgN8bL0Fyofee8IhN-0knxJF22/view"

alexart13 commented 5 years ago

@xuwei1245645564 This is a link to the model file on google drive. You need to download it if you want to use pretrained model.

xuwei1245645564 commented 5 years ago

@alexart13 Thanks. I know that this is a link ,but I don't it. Also, can you tell me the specific operation process of this project? Enter the size of the image, can I turn it up? Why didn't you choose the sequence in the original article for training?

look forward to your reply !

xuwei1245645564 commented 5 years ago

@alexart13 I have downloaded the pre-trained model. However, when I try to run "test,py" and "main.py",the error appears as follow: FileNotFoundError: [Errno 2] No such file or directory: '/media/xu/Elements/data/DeepVO-pytorch/models/t000102050809_v04060710_im184x608_s5x7_b4_rnn1000_optAdagrad_lr0.0005.model.train' You tell me that why?

ChiWeiHsiao commented 5 years ago

Hi @xuwei1245645564,

1) Yes, you can use larger image size, I reduce the image size just because memory issue. 2) Sorry I cannot recall precisely why I choose different sequences to the paper, but you can follow the original paper by specifying self.train_video and self.valid_video in params.py'. 3) Regarding to the error message, could you check if the downloaded model file is placed in the correct path?

xuwei1245645564 commented 5 years ago

@ChiWeiHsiao Thanks. One of the folders has the name: “datainfo”. What does it mean？If I use different sequences for training, will the files in this folder change? If you change the size of the input image, do you need to change the "self.img_means" and ‘’self.img_stds”? Is the path of this pre-training model not in this file "params.py"?