AttributeError: 'NoneType' object has no attribute 'siglip_embedder' #1246
Description
微调Z-Image，按照https://github.com/modelscope/DiffSynth-Studio/blob/main/docs/zh/Training/Split_Training.md的方法拆分成两阶段，在第一阶段生成cache时报错，报错内容如下：
2026-02-02 17:30:22,527 - modelscope - INFO - Download model 'Tongyi-MAI/Z-Image-Turbo' successfully.██████████████████████████████████████████████████████████████| 10.9M/10.9M [00:01<00:00, 11.0MB/s]
0%| | 0/106 [00:00<?, ?it/s]
Traceback (most recent call last):
File "/root/DiffSynth-Studio/examples/z_image/model_training/train.py", line 153, in
launcher_map[args.task](accelerator, dataset, model, model_logger, args=args)
File "/root/DiffSynth-Studio/diffsynth/diffusion/runner.py", line 70, in launch_data_process_task
data = model(data)
^^^^^^^^^^^
File "/root/miniconda3/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1773, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/root/miniconda3/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1784, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/root/DiffSynth-Studio/examples/z_image/model_training/train.py", line 87, in forward
inputs = self.pipe.unit_runner(unit, self.pipe, *inputs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/root/DiffSynth-Studio/diffsynth/diffusion/base_pipeline.py", line 435, in call
processor_outputs = unit.process(pipe, **processor_inputs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/root/DiffSynth-Studio/diffsynth/pipelines/z_image.py", line 299, in process
if hasattr(pipe, "dit") and pipe.dit.siglip_embedder is not None:
^^^^^^^^^^^^^^^^^^^^^^^^
AttributeError: 'NoneType' object has no attribute 'siglip_embedder'
下面是我的脚本：
accelerate launch DiffSynth-Studio/examples/z_image/model_training/train.py
--dataset_base_path autodl-tmp/dataset
--dataset_metadata_path autodl-tmp/dataset/metadata.csv
--max_pixels 1048576
--dataset_repeat 1
--model_paths '[
[
"autodl-tmp/z-image/text_encoder/model-00001-of-00003.safetensors",
"autodl-tmp/z-image/text_encoder/model-00002-of-00003.safetensors",
"autodl-tmp/z-image/text_encoder/model-00003-of-00003.safetensors"
],
"autodl-tmp/z-image/vae/diffusion_pytorch_model.safetensors"
]'
--learning_rate 1e-4
--num_epochs 5
--remove_prefix_in_ckpt "pipe.dit."
--output_path "autodl-tmp/z-image-cache"
--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
--task "sft:data_process"