Problems during sample about medsegdiff HOT 7 OPEN

Why is it that the result of my last training in BraTS is a MRI diagram?

from medsegdiff.

I run the segmentation_sample.py, and meet the problem:

Logging to /root/autodl-tmp/MedSegDif/med_results/img_out/ creating model and diffusion... sampling... no dpm-solver /root/miniconda3/lib/python3.8/site-packages/torch/nn/functional.py:1709: UserWarning: nn.functional.sigmoid is deprecated. Use torch.sigmoid instead. warnings.warn("nn.functional.sigmoid is deprecated. Use torch.sigmoid instead.") Traceback (most recent call last): File "MedSegDif/med_scripts/segmentation_sample.py", line 163, in main() File "MedSegDif/med_scripts/segmentation_sample.py", line 109, in main sample, x_noisy, org, cal, cal_out = sample_fn( File "/root/autodl-tmp/./MedSegDif/med_guided_diffusion/gaussian_diffusion.py", line 553, in p_sample_loop_known for sample in self.p_sample_loop_progressive( File "/root/autodl-tmp/./MedSegDif/med_guided_diffusion/gaussian_diffusion.py", line 624, in p_sample_loop_progressive out = self.p_sample( File "/root/autodl-tmp/./MedSegDif/med_guided_diffusion/gaussian_diffusion.py", line 435, in p_sample out = self.p_mean_variance( File "/root/autodl-tmp/./MedSegDif/med_guided_diffusion/respace.py", line 90, in p_mean_variance return super().p_mean_variance(self._wrap_model(model), *args, **kwargs) File "/root/autodl-tmp/./MedSegDif/med_guided_diffusion/gaussian_diffusion.py", line 319, in p_mean_variance model_mean, _, _ = self.q_posterior_mean_variance( File "/root/autodl-tmp/./MedSegDif/med_guided_diffusion/gaussian_diffusion.py", line 219, in q_posterior_mean_variance assert x_start.shape == x_t.shape AssertionError

I know that it is because x_start.shape is not equal to x_t.shape. However, My dataset is similar to ISICDataset, so I feel very strange. Thanks a lot if you can reply.

Have you solved it? I have the same doubt, thanks!

from medsegdiff.

Have you figure it out? I have the same question...

from medsegdiff.

could you print x_start.shape and x_t.shape?

from medsegdiff.

I try to print x_start.shape and x_t.shape however there is an error occurred before I can get x_start.

Traceback (most recent call last):
File "F:\dpm\MedSegDiff-master-v2\v2\scripts\segmentation_sample.py", line 194, in
main()
File "F:\dpm\MedSegDiff-master-v2\v2\scripts\segmentation_sample.py", line 116, in main
sample, x_noisy, org, cal, cal_out = sample_fn(
File "F:\dpm\MedSegDiff-master-v2\v2\scripts..\guided_diffusion\gaussian_diffusion.py", line 566, in p_sample_loop_known
for sample in self.p_sample_loop_progressive(
File "F:\dpm\MedSegDiff-master-v2\v2\scripts..\guided_diffusion\gaussian_diffusion.py", line 651, in p_sample_loop_progressive
out = self.p_sample(
File "F:\dpm\MedSegDiff-master-v2\v2\scripts..\guided_diffusion\gaussian_diffusion.py", line 446, in p_sample
out = self.p_mean_variance(
File "F:\dpm\MedSegDiff-master-v2\v2\scripts..\guided_diffusion\respace.py", line 90, in p_mean_variance
return super().p_mean_variance(self._wrap_model(model), *args, **kwargs)
File "F:\dpm\MedSegDiff-master-v2\v2\scripts..\guided_diffusion\gaussian_diffusion.py", line 325, in p_mean_variance
self._predict_xstart_from_eps(x_t=x, t=t, eps=model_output)
File "F:\dpm\MedSegDiff-master-v2\v2\scripts..\guided_diffusion\gaussian_diffusion.py", line 350, in _predict_xstart_from_eps
assert x_t.shape == eps.shape
AssertionError

I print x_t.shape and eps.shape, and result is like this:
x_t.shape= torch.Size([1, 1, 64, 64])
eps.shape= torch.Size([1, 2, 64, 64])

I suppose to use the right dataset step by step in Readme.md. The eps[:, 0, :, :] is totally the same as eps[:, 1; :, :]. After clipping eps to [1,1,64,64] by deleting either of them, program can work. I can't understand why their shapes are not consistant. Therefore, I can't figure out whether I can get the right answer by simple clipping conducted by myself without theory support.

from medsegdiff.

I try to print x_start.shape and x_t.shape however there is an error occurred before I can get x_start.

Traceback (most recent call last):
File "F:\dpm\MedSegDiff-master-v2\v2\scripts\segmentation_sample.py", line 194, in
main()
File "F:\dpm\MedSegDiff-master-v2\v2\scripts\segmentation_sample.py", line 116, in main
sample, x_noisy, org, cal, cal_out = sample_fn(
File "F:\dpm\MedSegDiff-master-v2\v2\scripts..\guided_diffusion\gaussian_diffusion.py", line 566, in p_sample_loop_known
for sample in self.p_sample_loop_progressive(
File "F:\dpm\MedSegDiff-master-v2\v2\scripts..\guided_diffusion\gaussian_diffusion.py", line 651, in p_sample_loop_progressive
out = self.p_sample(
File "F:\dpm\MedSegDiff-master-v2\v2\scripts..\guided_diffusion\gaussian_diffusion.py", line 446, in p_sample
out = self.p_mean_variance(
File "F:\dpm\MedSegDiff-master-v2\v2\scripts..\guided_diffusion\respace.py", line 90, in p_mean_variance
return super().p_mean_variance(self._wrap_model(model), *args, **kwargs)
File "F:\dpm\MedSegDiff-master-v2\v2\scripts..\guided_diffusion\gaussian_diffusion.py", line 325, in p_mean_variance
self._predict_xstart_from_eps(x_t=x, t=t, eps=model_output)
File "F:\dpm\MedSegDiff-master-v2\v2\scripts..\guided_diffusion\gaussian_diffusion.py", line 350, in _predict_xstart_from_eps
assert x_t.shape == eps.shape
AssertionError

I print x_t.shape and eps.shape, and result is like this: x_t.shape= torch.Size([1, 1, 64, 64]) eps.shape= torch.Size([1, 2, 64, 64])

I suppose to use the right dataset step by step in Readme.md. The is totally the same as . After clipping eps to by deleting either of them, program can work. I can't understand why their shapes are not consistant. Therefore, I can't figure out whether I can get the right answer by simple clipping conducted by myself without theory support.eps[:, 0, :, :]``eps[:, 1; :, :]``[1,1,64,64]

Hello! I have encountered a type-related issue with the ISIC dataset. Specifically, I'm getting the following error:

"assert x_t.shape == eps.shape
AssertionError"

Could you please help me identify which line of code needs to be modified to make the project run successfully?

from medsegdiff.

Has it been resolved? I encountered the same problem

from medsegdiff.