In my understanding,num_classes represents the number of categories that classifier can output,and 256 seems a little bit big.
I only change the config/gaussian_dataset/train.json num_classes from 256 to 128,and run train script,which outputs some errors.How can I modify it?
Hope for your reply!
output information follows:
Optimizing output/lerf/ramen
Output folder: output/lerf/ramen [04/06 16:04:42]
Reading camera 135/135 [04/06 16:04:42]
Loading Training Cameras [04/06 16:04:42]
Loading Test Cameras [04/06 16:04:49]
Number of points at initialisation : 30076 [04/06 16:04:49]
Num classes: 128 [04/06 16:04:50]
Training progress: 0%| | 0/30000 [00:00<?, ?it/s]Traceback (most recent call last):
File "train.py", line 312, in <module>
training(lp.extract(args), op.extract(args), pp.extract(args), args.test_iterations, args.save_iterations, args.checkpoint_iterations, args.start_checkpoint, args.debug_from, args.use_wandb)
File "train.py", line 106, in training
loss = (1.0 - opt.lambda_dssim) * Ll1 + opt.lambda_dssim * (1.0 - ssim(image, gt_image)) + loss_obj
File "/data_ssd/w50032615/code/gaussian-grouping-main/utils/loss_utils.py", line 46, in ssim
window = window.cuda(img1.get_device())
RuntimeError: CUDA error: an illegal memory access was encountered
CUDA kernel errors might be asynchronously reported at some other API call,so the stacktrace below might be incorrect.
For debugging consider passing CUDA_LAUNCH_BLOCKING=1.
Thanks for your work!
In my understanding,num_classes represents the number of categories that classifier can output,and 256 seems a little bit big. I only change the config/gaussian_dataset/train.json num_classes from 256 to 128,and run train script,which outputs some errors.How can I modify it?
Hope for your reply!