Skip to content

Feature/0308 mlm dual #146

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

Open
maxx1031 wants to merge 3 commits into
base: main
Choose a base branch
from
Open

Feature/0308 mlm dual #146

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
5fef04a
dual imag v0
Mar 6, 2026
7456b97
implement MLLM multi-image implementation
Mar 8, 2026
5344b1a
update training data configs and continue MLM
Mar 8, 2026
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
1 change: 1 addition & 0 deletions .claude/tsc-cache/d2ffffd0-e71d-465f-b644-2132f4e62ff5/affected-repos.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
root
6 changes: 6 additions & 0 deletions .claude/tsc-cache/d2ffffd0-e71d-465f-b644-2132f4e62ff5/edited-files.log
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,6 @@
1772731018:/scratch3/f007yzf/repos/Step1X-Edit-clean/inference_v2.py:root
1772731024:/scratch3/f007yzf/repos/Step1X-Edit-clean/inference_v2.py:root
1772731034:/scratch3/f007yzf/repos/Step1X-Edit-clean/inference_v2.py:root
1772731040:/scratch3/f007yzf/repos/Step1X-Edit-clean/inference_v2.py:root
1772731844:/scratch3/f007yzf/repos/Step1X-Edit-clean/inference_v2.py:root
1772731852:/scratch3/f007yzf/repos/Step1X-Edit-clean/inference_v2.py:root
11 changes: 10 additions & 1 deletion .gitignore
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,4 +3,13 @@
*/.DS_store
__pycache__
*/__pycache__/
test*
test*
tmp_*
training_data/source_img
training_data/reference_img
training_data/target_img
training_data/reference_img/left
training_data/reference_img/right
training_6k/source_img
training_6k/reference_img
training_6k/target_img
Binary file added examples/reference/0000_I.png
View file
Open in desktop
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file added examples/source/0000_l.png
View file
Open in desktop
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
6 changes: 3 additions & 3 deletions finetuning.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -143,7 +143,7 @@ def get_tokenize_strategy(self, args):
"""
获取分词策略。
"""
return strategy_step1x.Step1xEditTokenizeStrategy(tokenizer_cache_dir=args.qwen2p5vl)
return strategy_step1x.Step1xEditTokenizeStrategy(max_length=1280, tokenizer_cache_dir=args.qwen2p5vl)

def get_tokenizers(self, tokenize_strategy):
return [tokenize_strategy.processor]
Expand All @@ -157,7 +157,7 @@ def get_latents_caching_strategy(self, args):
return latents_caching_strategy

def get_text_encoding_strategy(self, args):
return strategy_step1x.Step1XEditEncodingStrategy()
return strategy_step1x.Step1XEditEncodingStrategy(max_length=1280)

def post_process_network(self, args, accelerator, network, text_encoders, unet):
pass
Expand Down Expand Up @@ -556,4 +556,4 @@ def setup_parser() -> argparse.ArgumentParser:
args = train_util.read_config_from_file(args, parser)

trainer = Step1XEditNetworkTrainer()
trainer.train(args)
trainer.train(args)
Loading