star
star
Moreover, in the `run_batch.py`, the pad settings are different among these scripts (0 or 24). I cannot reproduce the accuracy because the `pad` is 24 in 736d9a4 version, while the...
Have you fixed the issue, I have met the same problem.
Hi，I have found the issue caused by the version 1.4.1. I reinstall the pytorch-lightning==1.3.5 which have fixed the issue follow the command: `pip install pytorch-lightning==1.3.5`
@farhansajid Did you fix it? Does it work(python3.x)? Can you share me a copy?
> @foocross As I said, the dataset is too **bad**, so COLMAP can not match the images. So ran the three commands I mentioned to see if your images are...
> I have fix the issue at https://github.com/Fyusion/LLFF/pull/60 The issue is because some views(images) are not registered in the colmap, so we cannot use pose_boundary.npy to process all views(images)
> > I've just spent half an hour on google COFW and 300W images, without success. Wouldn't it be easier just include those official links? > > The 300W dataset...
> > > 转之前 .meta文件有140M，.data-00000有8.7M，应该是正常的 > > > > > > `def freeze_graph(input_checkpoint,output_graph): > > ``` > > # checkpoint = tf.train.get_checkpoint_state(model_folder) > > # input_checkpoint = checkpoint.model_checkpoint_path > >...
> It really works, but why? CUDA is avaliable for that situation