Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

high resolution #11

Open
wants to merge 5 commits into
base: main
Choose a base branch
from
Open

high resolution #11

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
e8d4a48
add quite a lot
ClashLuke Jul 9, 2024
d490aee
sync config attributes
ClashLuke Jul 9, 2024
4461d1a
fix(generator): move batch size computation
ClashLuke Jul 9, 2024
34ed4e8
more args
ClashLuke Jul 9, 2024
87897b9
perf(generator): don't use null prompt
ClashLuke Jul 16, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
5 changes: 4 additions & 1 deletion fabric/configs/preference_model_feedback.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -5,6 +5,7 @@ hydra:
job:
chdir: true

size: 512
model_version: "1.5"
model_name: dreamlike-art/dreamlike-photoreal-2.0
negative_prompt: lowres, bad anatomy, bad hands, cropped, worst quality
Expand All @@ -23,11 +24,13 @@ feedback:
min_weight: 0.0
max_weight: 0.3
neg_scale: 0.5
warmup_power: 1
global_seed: 0
image_seed: null
liked_images: []
disliked_images: []
output_path: images

# For loading the Human Preference Score LoRA weights, set null to not use them
lora_weights: null # resources/hps_lora/adapted_model.bin
lora_weights: null # resources/hps_lora/adapted_model.bin
topk: 0
17 changes: 10 additions & 7 deletions fabric/configs/single_round.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,17 +1,20 @@
model_version: "1.5"
model_ckpt: null
model_name: dreamlike-art/dreamlike-photoreal-2.0
prompt: photo of a dog running on grassland, masterpiece, best quality, fine details
prompt: A robot holding a sign that says "The future is now", masterpiece, best quality
negative_prompt: lowres, bad anatomy, bad hands, cropped, worst quality
liked: []
disliked: []
n_images: 4
denoising_steps: 20
guidance_scale: 7
n_images: 1
denoising_steps: 100
guidance_scale: 6
seed: 37
feedback:
start: 0.33
end: 0.66
start: 0.0
end: 0.8
min_weight: 0.05
max_weight: 0.8
max_weight: 0.9
neg_scale: 0.5
warmup_power: 1
topk: 4
size: 512
7 changes: 5 additions & 2 deletions fabric/configs/target_image_feedback.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -5,13 +5,14 @@ hydra:
job:
chdir: true

size: 512
model_version: "1.5"
model_name: dreamlike-art/dreamlike-photoreal-2.0 # runwayml/stable-diffusion-v1-5
negative_prompt: lowres, bad anatomy, bad hands, cropped, worst quality
n_images: 4
n_rounds: 3
num_prompts: 1000
denoising_steps: 20
denoising_steps: 10
guidance_scale: 6
prompt_dropout: 0.0
use_pos_feedback: yes
Expand All @@ -22,6 +23,7 @@ feedback:
min_weight: 0.0
max_weight: 0.3
neg_scale: 0.5
warmup_power: 1
global_seed: 0
image_seed: null
liked_images: []
Expand All @@ -30,4 +32,5 @@ prompthero_path: resources/prompthero
output_path: images

# For loading the Human Preference Score LoRA weights, set null to not use them
lora_weights: null # resources/hps_lora/adapted_model.bin
lora_weights: null # resources/hps_lora/adapted_model.bin
topk: 0
Loading