Training Example

1. training example for t2i sft or lora

• Data format

You need to create a jsonl file with key-values in the table below:

key_word	Required	Description	Example
img_path	Required	image path	./data_example/images/0.png
prompt	Required	text	A lovely little girl.
width	Required	image width	1024
height	Required	image height	1024

• Tainging Scripts

bash ./train_examples/sft/train.sh

# All training setting in train_config.yaml

# --data_csv_root:  data csv_filepath
# --aspect_ratio_type: data bucketing strategy, mar_256、mar_512、mar_1024
# --pretrained_model_name_or_path: root directory of the model
# --diffusion_pretrain_weight: if a specified diffusion weight path is provided, load the model parameters from the current directory.
# --work_dir: the save root directory for ckpt and logs
# --resume_from_checkpoint: If 'resume_from_checkpoint' is set to 'latest', load the most recent step checkpoint. If a specific directory is provided, resume training from that directory.

2. training example for t2i dpo

• Data format

You need to create a txt file with key-values in the table below:

key_word	Required	Description	Example
img_path_win	Required	win image path	./data_example/images/0.png
img_path_lose	Required	lose image path	./data_example/images/1.png
prompt	Required	text	A lovely little girl.
width	Required	image width	1024
height	Required	image height	1024

• Tainging Scripts

bash ./train_examples/dpo/train.sh

# All training setting in train_config.yaml

# --data_txt_root:  data txt_filepath
# --aspect_ratio_type: data bucketing strategy, mar_256、mar_512、mar_1024
# --pretrained_model_name_or_path: root directory of the model
# --diffusion_pretrain_weight: if a specified diffusion weight path is provided, load the model parameters from the current directory.
# --work_dir: the save root directory for ckpt and logs
# --resume_from_checkpoint: If 'resume_from_checkpoint' is set to 'latest', load the most recent step checkpoint. If a specific directory is provided, resume training from that directory.

3. training example for edit sft or lora

• Data format

You need to create a txt file with key-values in the table below:

key_word	Required	Description	Example
img_path	Required	edited image path	./data_example/images/0_edited.png
ref_img_path	Required	raw image path	./data_example/images/0.png
prompt	Required	edit instruction	change the dog to cat.
width	Required	image width	1024
height	Required	image height	1024

• Tainging Scripts

bash ./train_examples/edit_sft/train.sh

# All training setting in train_config.yaml

# --data_txt_root:  data txt_filepath
# --aspect_ratio_type: data bucketing strategy, mar_256、mar_512、mar_1024
# --pretrained_model_name_or_path: root directory of the model
# --diffusion_pretrain_weight: if a specified diffusion weight path is provided, load the model parameters from the current directory.
# --work_dir: the save root directory for ckpt and logs
# --resume_from_checkpoint: If 'resume_from_checkpoint' is set to 'latest', load the most recent step checkpoint. If a specific directory is provided, resume training from that directory.

4. training example for edit dpo

• Data format

You need to create a txt file with key-values in the table below:

key_word	Required	Description	Example
img_path_win	Required	win image path	./data_example/images/0_win.png
img_path_lose	Required	lose image path	./data_example/images/0_lose.png
ref_img_path	Required	ref image path	./data_example/images/0_ref.png
prompt	Required	text	change the dog to cat.
width	Required	image width	1024
height	Required	image height	1024

• Tainging Scripts

bash ./train_examples/edit_dpo/train.sh

# All training setting in train_config.yaml

# --data_txt_root:  data txt_filepath
# --aspect_ratio_type: data bucketing strategy, mar_256、mar_512、mar_1024
# --pretrained_model_name_or_path: root directory of the model
# --diffusion_pretrain_weight: if a specified diffusion weight path is provided, load the model parameters from the current directory.
# --work_dir: the save root directory for ckpt and logs
# --resume_from_checkpoint: If 'resume_from_checkpoint' is set to 'latest', load the most recent step checkpoint. If a specific directory is provided, resume training from that directory.