Song Luo

Parameter Size Increase

2

Hi, thanks for the great implementation of tensor decomposition. However, when I apply your code to my resnet model, the decomposed model actually has more parameters than the original one....

MAE using pretrained VIT

3

Hi There, I am currently trying to fine-tune an MAE based on pretrained VIT from timm. However, when I do: ``` v = timm.create_model('vit_base_patch16_224', pretrained=True) num_ftrs = v.head.in_features v.head =...

[Bug] CG terminated in 1000 iterations with average residual norm XXX which is larger than the tolerance of 1 specified by gpytorch.settings.cg_tolerance.

1

# 🐛 Bug I am not sure this is a bug or something I just made a mistake. I do notice there was a post (issue #1129) very similar to...

Is there any way I can test using CPU

8

Several issues here. After I torch.cuda.set_device(-1) segmentation_module.cuda() and test(segmentation_module, loader_test, -1) do not perform well. Is there any way I can test using CPU?