search

researchmm / AOT-GAN-for-Inpainting

[TVCG'2023] AOT-GAN for High-Resolution Image Inpainting (codebase for image inpainting)
https://arxiv.org/abs/2104.01431
Apache License 2.0
403 stars 64 forks source link

Fixed conda environment error and 'num_samples=0' error #10

Open Dutra-Apex opened 1 year ago

Dutra-Apex commented 1 year ago

When trying to create a conda venv, I kept getting errors about the libraries not being found. To fix this I removed the build string from the environment.yml file. Code ran fine despite the change.

The second error was 'num_samples=0', and it happened when trying to run 'train.py' while passing a custom image and mask directory. The error was fixed by changing how the image_path and mask_path are created.

Hope this is helpful to anyone that might get the same problems.

xinzihao commented 8 months ago

D:\anaconda\envs\torch\python.exe D:\pythonProject12\workplace\AOT-GAN-for-Inpainting-master\AOT-GAN-for-Inpainting-master\src\train.py [**] create folder ../experiments\aotgan_places2_pconv256 D:\anaconda\envs\torch\lib\site-packages\torchvision\models_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and will be removed in 0.15, please use 'weights' instead. warnings.warn( D:\anaconda\envs\torch\lib\site-packages\torchvision\models_utils.py:223: UserWarning: Arguments other than a weight enum or None for 'weights' are deprecated since 0.13 and will be removed in 0.15. The current behavior is equivalent to passing weights=VGG19_Weights.IMAGENET1K_V1. You can also use weights=VGG19_Weights.DEFAULT to get the most up-to-date weights. warnings.warn(msg) 0%| | 0/10 [00:00<?, ?it/s] @Dutra-Apex May I ask how to solve this problem，thanks.