wwhio comments

Results 6 comments of


                                            wwhio

call `deform_conv2d` from torchvision when possible

Thank you for your reply. I knew that. But my PR will work with all DCN related things instead of DCNv2Pack only. And I think this could be a solution...

first stage of train.py for ESRGAN is slow and single threaded

IMO, the time consuming operation is here: https://github.com/xinntao/BasicSR/blob/5c757162b348a09d236e00c2cc04463c0a8bba45/basicsr/data/data_sampler.py#L33 The operation here is intended to keep reproducibility. If you have a large dataset, try to reduce `dataset_enlarge_ratio` in the configuration file...

from torch.optim.lr_scheduler import _LRScheduler 报错

`_LRScheduler` 是 Pytorch 提供的各种 LR Scheduler 的基类，请检查： 1. 是否在编辑器 / IDE 中选择了正确的 Python 解释器 2. 是否正确安装了 pytorch 参考： https://pytorch.org/docs/stable/_modules/torch/optim/lr_scheduler.html

数据集 gt是什么意思

GT (ground truth) 在这里是参考图像。对应 LQ 是参考高质量的图像，对应 LR 是参考的高分辨率（HR）图像，其他类推。

数据集 gt是什么意思

是。 LQ ： Low Quality, e.g. LR (Low Resolution), blurry, noisy, etc

Question [ How configure the plugin with venvs ]

1. activate your venv in a terminal 2. start lapce in the same terminal