Skip to content
GitLab
Menu
Projects
Groups
Snippets
Loading...
Help
Help
Support
Community forum
Keyboard shortcuts
?
Submit feedback
Contribute to GitLab
Sign in
Toggle navigation
Menu
Open sidebar
chenpangpang
diffusers
Commits
23a2cd33
Unverified
Commit
23a2cd33
authored
Jun 18, 2024
by
Sayak Paul
Committed by
GitHub
Jun 18, 2024
Browse files
[LoRA] training fix the position of param casting when loading them (#8460)
fix the position of param casting when loading them
parent
4edde134
Changes
2
Hide whitespace changes
Inline
Side-by-side
Showing
2 changed files
with
4 additions
and
4 deletions
+4
-4
examples/dreambooth/train_dreambooth_lora_sdxl.py
examples/dreambooth/train_dreambooth_lora_sdxl.py
+2
-2
examples/research_projects/scheduled_huber_loss_training/dreambooth/train_dreambooth_lora_sdxl.py
...er_loss_training/dreambooth/train_dreambooth_lora_sdxl.py
+2
-2
No files found.
examples/dreambooth/train_dreambooth_lora_sdxl.py
View file @
23a2cd33
...
@@ -1289,8 +1289,8 @@ def main(args):
...
@@ -1289,8 +1289,8 @@ def main(args):
models
=
[
unet_
]
models
=
[
unet_
]
if
args
.
train_text_encoder
:
if
args
.
train_text_encoder
:
models
.
extend
([
text_encoder_one_
,
text_encoder_two_
])
models
.
extend
([
text_encoder_one_
,
text_encoder_two_
])
# only upcast trainable parameters (LoRA) into fp32
# only upcast trainable parameters (LoRA) into fp32
cast_training_params
(
models
)
cast_training_params
(
models
)
accelerator
.
register_save_state_pre_hook
(
save_model_hook
)
accelerator
.
register_save_state_pre_hook
(
save_model_hook
)
accelerator
.
register_load_state_pre_hook
(
load_model_hook
)
accelerator
.
register_load_state_pre_hook
(
load_model_hook
)
...
...
examples/research_projects/scheduled_huber_loss_training/dreambooth/train_dreambooth_lora_sdxl.py
View file @
23a2cd33
...
@@ -1363,8 +1363,8 @@ def main(args):
...
@@ -1363,8 +1363,8 @@ def main(args):
models
=
[
unet_
]
models
=
[
unet_
]
if
args
.
train_text_encoder
:
if
args
.
train_text_encoder
:
models
.
extend
([
text_encoder_one_
,
text_encoder_two_
])
models
.
extend
([
text_encoder_one_
,
text_encoder_two_
])
# only upcast trainable parameters (LoRA) into fp32
# only upcast trainable parameters (LoRA) into fp32
cast_training_params
(
models
)
cast_training_params
(
models
)
accelerator
.
register_save_state_pre_hook
(
save_model_hook
)
accelerator
.
register_save_state_pre_hook
(
save_model_hook
)
accelerator
.
register_load_state_pre_hook
(
load_model_hook
)
accelerator
.
register_load_state_pre_hook
(
load_model_hook
)
...
...
Write
Preview
Markdown
is supported
0%
Try again
or
attach a new file
.
Attach a file
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to comment