aboutsummaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorXavierXiao <xiaozhisheng950@gmail.com>2022-09-06 09:41:12 -0700
committerXavierXiao <xiaozhisheng950@gmail.com>2022-09-06 09:41:12 -0700
commitd5def4abd8cde2b5c3f6c53643fa53a27c175aa7 (patch)
tree416137660e3b189bef632dd7bb9c472423c3fea8
parentchange config (diff)
parentUpdate README.md (diff)
Merge branch 'main' of github.com:XavierXiao/Dreambooth-Stable-Diffusion
-rw-r--r--README.md4
1 files changed, 4 insertions, 0 deletions
diff --git a/README.md b/README.md
index 7133ca3..2e0ac8d 100644
--- a/README.md
+++ b/README.md
@@ -33,4 +33,8 @@ python main.py --base configs/stable-diffusion/v1-finetune_unfrozen.yaml
--class_word <xxx>
```
+Detailed configuration can be found in ```configs/stable-diffusion/v1-finetune_unfrozen.yaml```. In particular, the default learning rate is ```1.0e-6``` as I found the ```1.0e-5``` in the Dreambooth paper leads to poor editability. The parameter ```reg_weight``` corresponds to the weight of regularization in the Dreambooth paper, and the default is set to ```1.0```.
+
+Dreambooth requires a placeholder word ```[V]```, called identifier, as in the paper. This identifier needs to be a relatively rare tokens in the vocabulary. The original paper approaches this by using a rare word in T5-XXL tokenizer. For simplicity, here I just use a random word ```sks``` and hard coded it.. If you want to change that, simply make a change in [this file](https://github.com/XavierXiao/Dreambooth-Stable-Diffusion/blob/main/ldm/data/personalized.py#L10).
+
### Generation