theluyuan/DiffSynth-Studio
1
0
Fork 0
mirror of https://github.com/modelscope/DiffSynth-Studio.git synced 2026-03-23 00:58:11 +00:00
Code Issues Packages Projects Releases Wiki Activity

support z-image controlnet

Browse Source
This commit is contained in:
Artiprocher
2026-01-07 15:56:53 +08:00
parent 32449a6aa0
commit bac39b1cd2
28 changed files with 868 additions and 11 deletions
Download Patch File Download Diff File Expand all files Collapse all files

20
examples/z_image/model_training/lora/Z-Image-Omni-Base.sh
View File

@@ -1,3 +1,4 @@
# Text to image training
accelerate launch examples/z_image/model_training/train.py \
--dataset_base_path data/example_image_dataset \
--dataset_metadata_path data/example_image_dataset/metadata.csv \
@@ -13,3 +14,22 @@ accelerate launch examples/z_image/model_training/train.py \
--lora_rank 32 \
--use_gradient_checkpointing \
--dataset_num_workers 8
# Image(s) to image training
# accelerate launch examples/z_image/model_training/train.py \
# --dataset_base_path data/example_image_dataset \
# --dataset_metadata_path data/example_image_dataset/metadata_qwen_imgae_edit_multi.json \
# --data_file_keys "image,edit_image" \
# --extra_inputs "edit_image" \
# --max_pixels 1048576 \
# --dataset_repeat 50 \
# --model_id_with_origin_paths "Tongyi-MAI/Z-Image-Omni-Base:transformer/*.safetensors,Tongyi-MAI/Z-Image-Omni-Base:siglip/model.safetensors,Tongyi-MAI/Z-Image-Turbo:text_encoder/*.safetensors,Tongyi-MAI/Z-Image-Turbo:vae/diffusion_pytorch_model.safetensors" \
# --learning_rate 1e-4 \
# --num_epochs 5 \
# --remove_prefix_in_ckpt "pipe.dit." \
# --output_path "./models/train/Z-Image-Omni-Base_lora_edit" \
# --lora_base_model "dit" \
# --lora_target_modules "to_q,to_k,to_v,to_out.0,w1,w2,w3" \
# --lora_rank 32 \
# --use_gradient_checkpointing \
# --dataset_num_workers 8

17
examples/z_image/model_training/lora/Z-Image-Turbo-Fun-Controlnet-Tile-2.1-8steps.sh Normal file
View File

@@ -0,0 +1,17 @@
accelerate launch examples/z_image/model_training/train.py \
--dataset_base_path data/example_image_dataset \
--dataset_metadata_path data/example_image_dataset/metadata_controlnet_upscale.csv \
--data_file_keys "image,controlnet_image" \
--max_pixels 1048576 \
--dataset_repeat 100 \
--model_id_with_origin_paths "PAI/Z-Image-Turbo-Fun-Controlnet-Union-2.1:Z-Image-Turbo-Fun-Controlnet-Tile-2.1-8steps.safetensors,Tongyi-MAI/Z-Image-Turbo:transformer/*.safetensors,Tongyi-MAI/Z-Image-Turbo:text_encoder/*.safetensors,Tongyi-MAI/Z-Image-Turbo:vae/diffusion_pytorch_model.safetensors" \
--learning_rate 1e-4 \
--num_epochs 5 \
--remove_prefix_in_ckpt "pipe.dit." \
--output_path "./models/train/Z-Image-Turbo-Fun-Controlnet-Tile-2.1-8steps_lora" \
--lora_base_model "dit" \
--lora_target_modules "to_q,to_k,to_v,to_out.0,w1,w2,w3" \
--lora_rank 32 \
--extra_inputs "controlnet_image" \
--use_gradient_checkpointing \
--dataset_num_workers 8

17
examples/z_image/model_training/lora/Z-Image-Turbo-Fun-Controlnet-Union-2.1-8steps.sh Normal file
View File

@@ -0,0 +1,17 @@
accelerate launch examples/z_image/model_training/train.py \
--dataset_base_path data/example_image_dataset \
--dataset_metadata_path data/example_image_dataset/metadata_controlnet_canny.csv \
--data_file_keys "image,controlnet_image" \
--max_pixels 1048576 \
--dataset_repeat 100 \
--model_id_with_origin_paths "PAI/Z-Image-Turbo-Fun-Controlnet-Union-2.1:Z-Image-Turbo-Fun-Controlnet-Union-2.1-8steps.safetensors,Tongyi-MAI/Z-Image-Turbo:transformer/*.safetensors,Tongyi-MAI/Z-Image-Turbo:text_encoder/*.safetensors,Tongyi-MAI/Z-Image-Turbo:vae/diffusion_pytorch_model.safetensors" \
--learning_rate 1e-4 \
--num_epochs 5 \
--remove_prefix_in_ckpt "pipe.dit." \
--output_path "./models/train/Z-Image-Turbo-Fun-Controlnet-Union-2.1-8steps_lora" \
--lora_base_model "dit" \
--lora_target_modules "to_q,to_k,to_v,to_out.0,w1,w2,w3" \
--lora_rank 32 \
--extra_inputs "controlnet_image" \
--use_gradient_checkpointing \
--dataset_num_workers 8

17
examples/z_image/model_training/lora/Z-Image-Turbo-Fun-Controlnet-Union-2.1.sh Normal file
View File

@@ -0,0 +1,17 @@
accelerate launch examples/z_image/model_training/train.py \
--dataset_base_path data/example_image_dataset \
--dataset_metadata_path data/example_image_dataset/metadata_controlnet_canny.csv \
--data_file_keys "image,controlnet_image" \
--max_pixels 1048576 \
--dataset_repeat 100 \
--model_id_with_origin_paths "PAI/Z-Image-Turbo-Fun-Controlnet-Union-2.1:Z-Image-Turbo-Fun-Controlnet-Union-2.1.safetensors,Tongyi-MAI/Z-Image-Turbo:transformer/*.safetensors,Tongyi-MAI/Z-Image-Turbo:text_encoder/*.safetensors,Tongyi-MAI/Z-Image-Turbo:vae/diffusion_pytorch_model.safetensors" \
--learning_rate 1e-4 \
--num_epochs 5 \
--remove_prefix_in_ckpt "pipe.dit." \
--output_path "./models/train/Z-Image-Turbo-Fun-Controlnet-Union-2.1_lora" \
--lora_base_model "dit" \
--lora_target_modules "to_q,to_k,to_v,to_out.0,w1,w2,w3" \
--lora_rank 32 \
--extra_inputs "controlnet_image" \
--use_gradient_checkpointing \
--dataset_num_workers 8