Downloads · 30 days
0
controlnet-channels
GeroldMeisinger/controlnet-channels
controlnet-channels is a machine learning model from GeroldMeisinger. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as cc-by-nc-sa-4.0.
Access
Public
Updated Oct 25, 2023
Repo size
749 MB
Likes
1
Public
Repo makeup
Click a slice to open those files.
.safetensors723 MB · 97%
At a glance
- License
- cc-by-nc-sa-4.0
- Access
- Public
- Created
- Oct 25, 2023
- Updated
- Oct 25, 2023
- SHA
- f6fba872
Model card
From the Hugging Face model README
Restore missing RGB channels
Restore a missing channel of a RGB image by using ControlNet to guide image generation of Stable Diffusion to infer missing channel from the other two channels.
- See accompanying discussion at github.com - Channels RGB with detailed report and evaluations.
- To restore images with missing channels you can use this space.
- For evaluation images see the corresponding .zip's at "files".
- To run your own evaluations you can use this script at gitlab.com.
Training
accelerate launch train_controlnet.py \
--pretrained_model_name_or_path="runwayml/stable-diffusion-v1-5" \
--train_batch_size=4 \
--gradient_accumulation_steps=8 \
--proportion_empty_prompts=0.5
--mixed_precision="fp16" \
--learning_rate=1e-5 \
--enable_xformers_memory_efficient_attention \
--use_8bit_adam \
--set_grads_to_none \
--seed=0 \
--num_train_epochs=2
Image dataset
- laion2B-en aesthetics>=6.5 dataset
- --min_image_size 512 --max_aspect_ratio 2 --resize_mode="center_crop" --image_size 512
- Cleaned with
fastdupdefault settings - Data augmented with right-left flipped images
- Resulting in 214244 images
- Set whole channel to 0 by alternating between R-G-B channels
- License
- cc-by-nc-sa-4.0
- Created
- Oct 25, 2023
- Updated
- Oct 25, 2023