nodes/Extraltodeus-DistanceSampler
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Update sampler_explaination.md

Browse Source
This commit is contained in:
Extraltodeus
2025-04-27 04:14:51 +02:00
committed by GitHub
parent e24405543b
commit 3055347392
1 changed files with 5 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files
sampler_explaination.md
+5 -1
View File
@@ -64,7 +64,11 @@ If an unconditional tensor uncond is provided (which is relevant for Classifier-
uncond = uncond.div(torch.linalg.matrix_norm(uncond, keepdim=True))
Then, the distance of each normalized tensor tn from the normalized unconditional tensor is calculated and added to the existing distances: distances \+= tn.sub(uncond).abs().div(n). This step incorporates information about how far each conditional sample is from the unconditional sample into the weighting scheme.
Then, the distance of each normalized tensor tn from the normalized unconditional tensor is calculated and added to the existing distances:
distances += tn.sub(uncond).abs().div(n)
This step incorporates information about how far each conditional sample is from the unconditional sample into the weighting scheme.
The function then proceeds to normalize the distances to obtain weights. If use\_softmax is True, the softmax function is applied: