gaozhihan / gaozhihan/PreDiff

LDM generate full of Noise

Open
#40 1 comment 0 reactions 1 assignee Claimed by @gaozhihan View on GitHub
question
Dominant language
Python
Stars
159
Forks
10
PR merge metrics
No merged PRs in 30d

Description

I use your model and train on my own dataset. My Dataset contains about 4000 samples for training. I do not know whether it is enough for training or not. How many samples do you have for training? The VAE can reconstruct the image in a perfect way. But the LDM cannot generate a meaningfull image.
This is an example in training set at epoch 80.
Image

What happen with me ?

Contributor guide

No contributing guide indexed for this repository

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.