Image-to-Image
English
stable-diffusion
controlnet
File size: 1,104 Bytes
e5214f3
 
109b8ce
 
 
e9ccc1c
109b8ce
c8e30b1
 
419de76
 
 
60ae6f9
 
 
 
 
 
 
 
 
591d167
60ae6f9
 
 
e9ccc1c
60ae6f9
 
 
 
 
e9ccc1c
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
---
license: mit
datasets:
- Levi980623/ControlNet_ade20k
- mfidabel/sam-coyo-3k
- lmms-lab/COCO-Caption2017
pipeline_tag: image-to-image
language:
- en
tags:
- stable-diffusion
- controlnet
---
# Model Card for lavinal712/controlnet-v1-seg

## Model Description

This is controlnet weight trained on runwayml/stable-diffusion-v1-5 with segmentaion.

### Training

This model was trained using a Segmented dataset based on the ADE20k Dataset, COYO-700M Dataset and COCO Dataset. Stable Diffusion v1.5 checkpoint was used as the base model for the controlnet.

- [ControlNet-ade20k](https://huggingface.co/datasets/Levi980623/ControlNet_ade20k)
- [SAM-COYO-3k](https://huggingface.co/datasets/mfidabel/sam-coyo-3k)
- [COCO2017](https://huggingface.co/datasets/lmms-lab/COCO-Caption2017)

#### Training Method

- Train on [ControlNet-ade20k](https://huggingface.co/datasets/Levi980623/ControlNet_ade20k) for 30 epochs.
- Train on [SAM-COYO-3k](https://huggingface.co/datasets/mfidabel/sam-coyo-3k) for 30 epochs.
- Train on [COCO2017](https://huggingface.co/datasets/lmms-lab/COCO-Caption2017) for 1 epoch.