[Flux] Dreambooth LoRA training scripts #9086
sayakpaul
merged
59 commits into
huggingface:main
from
linoytsaban:flux-fine-tuning
initial commit - dreambooth for flux
fb5488d9
update transformer to be FluxTransformer2DModel
3062dafc
update training loop and validation inference
1448aa1f
fix sd3->flux docs
a59e0120
Merge branch 'huggingface:main' into flux-fine-tuning
b2252834
add guidance handling, not sure if it makes sense(?)
df074d43
Merge remote-tracking branch 'origin/flux-fine-tuning' into flux-fine…
4f7c88c3
inital dreambooth lora commit
c897c31f
fix text_ids in compute_text_embeddings
c3c38a42
Merge branch 'main' into flux-fine-tuning
5e92ba94
linoytsaban
changed the title 
bghira
commented
on 2024-08-06
bghira
commented
on 2024-08-06
bghira
commented
on 2024-08-06
bghira
commented
on 2024-08-06
bghira
commented
on 2024-08-06
bghira
commented
on 2024-08-06
bghira
commented
on 2024-08-06
bghira
commented
on 2024-08-06
bghira
commented
on 2024-08-06
bghira
commented
on 2024-08-06
bghira
commented
on 2024-08-06
Merge branch 'main' into flux-fine-tuning
407459e4
fix imports of static methods
259e4433
Merge remote-tracking branch 'origin/flux-fine-tuning' into flux-fine…
f3f3c7ba
fix pipeline loading in readme, remove auto1111 docs for now
3d5b713b
fix pipeline loading in readme, remove auto1111 docs for now, remove …
cdbb69c6
Update examples/dreambooth/train_dreambooth_flux.py
b249d36d
fix te2 loading and remove te2 refs from text encoder training
37b8e645
fix tokenizer_2 initialization
0714278f
remove text_encoder training refs from lora script (for now)
80c3fe04
try with vae in bfloat16, fix model hook save
77a02358
Merge branch 'main' into flux-fine-tuning
2fcba983
fix tokenization
a64f4a52
Merge remote-tracking branch 'origin/flux-fine-tuning' into flux-fine…
18a23d45
fix static imports
08a12968
fix CLIP import
e4746830
remove text_encoder training refs (for now) from lora script
97f15460
fix minor bug in encode_prompt, add guidance def in lora script, ...
187e4217
fix unpack_latents args
b24f6732
Merge branch 'main' into flux-fine-tuning
72cb8596
fix license in readme
ad1d2368
add "none" to weighting_scheme options for uniform sampling
bcb752ba
style
df880f3c
adapt model saving - remove text encoder refs
e69244ad
adapt model loading - remove text encoder refs
66125321
initial commit for readme
155dbb2f
sayakpaul
commented
on 2024-08-08
sayakpaul
commented
on 2024-08-08
sayakpaul
commented
on 2024-08-08
sayakpaul
commented
on 2024-08-08
sayakpaul
commented
on 2024-08-08
sayakpaul
commented
on 2024-08-08
sayakpaul
commented
on 2024-08-08
sayakpaul
commented
on 2024-08-08
sayakpaul
commented
on 2024-08-08
sayakpaul
commented
on 2024-08-08
sayakpaul
commented
on 2024-08-08
sayakpaul
commented
on 2024-08-08
sayakpaul
commented
on 2024-08-08
sayakpaul
commented
on 2024-08-08
Merge branch 'main' into flux-fine-tuning
c7ea6204
Update examples/dreambooth/train_dreambooth_lora_flux.py
15261410
Update examples/dreambooth/train_dreambooth_lora_flux.py
6b78e197
fix vae casting
a2ac0eb8
remove precondition_outputs
7f0fe8a8
readme
d0fb7277
readme
dcd26d1f
style
60c5b650
linoytsaban
commented
on 2024-08-08
readme
aac9183b
Merge remote-tracking branch 'origin/flux-fine-tuning' into flux-fine…
0ebc3dd3
readme
56059d82
update weighting scheme default & docs
911306aa
style
8e4d2300
linoytsaban
commented
on 2024-08-08
add text_encoder training to lora script, change vae_scale_factor val…
573026c2
style
aea5d1fb
text encoder training fixes
bde7dedc
Merge remote-tracking branch 'origin/flux-fine-tuning' into flux-fine…
4e35d2b3
style
f7816303
Merge branch 'main' into flux-fine-tuning
5a86a554
update readme
d77b67f6
minor fixes
1d8e25f8
fix te params
dc1b10e3
fix te params
569f2e17
bghira
approved these changes
on 2024-08-08
Merge branch 'main' into flux-fine-tuning
c7097ab2
Login to write a write a comment.
Login via GitHub