RuntimeError: The size of tensor a (4) must match the size of tensor b (2) at non-singleton dimension 0

[gerylavin]

i don't know how to soolve this
Using decoupled weight decay
Downloading data: 63%|██████▎ | 17/27 [00:01<00:00, 20.77files/s]Downloading data: 100%|██████████| 27/27 [00:01<00:00, 30.40files/s]Downloading data: 100%|██████████| 27/27 [00:01<00:00, 21.50files/s]
Generating train split: 0%| | 0/26 [00:00<?, ? examples/s]Generating train split: 100%|██████████| 26/26 [00:00<00:00, 1906.90 examples/s]
Caching latents: 98%|█████████▊| 49/50 [00:12<00:00, 4.04it/s]Caching latents: 100%|██████████| 50/50 [00:12<00:00, 4.04it/s]Caching latents: 100%|██████████| 50/50 [00:12<00:00, 3.88it/s]
10/07/2025 16:36:22 - INFO - main - ***** Running training *****
10/07/2025 16:36:22 - INFO - main - Num examples = 100
10/07/2025 16:36:22 - INFO - main - Num batches each epoch = 50
10/07/2025 16:36:22 - INFO - main - Num Epochs = 20
10/07/2025 16:36:22 - INFO - main - Instantaneous batch size per device = 2
10/07/2025 16:36:22 - INFO - main - Total train batch size (w. parallel, distributed & accumulation) = 2
10/07/2025 16:36:22 - INFO - main - Gradient Accumulation steps = 1
10/07/2025 16:36:22 - INFO - main - Total optimization steps = 1000
Steps: 0%| | 0/1000 [00:00<?, ?it/s]Traceback (most recent call last):
File "/root/diffusers/examples/advanced_diffusion_training/train_dreambooth_lora_flux_advanced.py", line 2476, in
main(args)
File "/root/diffusers/examples/advanced_diffusion_training/train_dreambooth_lora_flux_advanced.py", line 2262, in main
(weighting.float() * (model_pred_prior.float() - target_prior.float()) ** 2).reshape(
~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
RuntimeError: The size of tensor a (4) must match the size of tensor b (2) at non-singleton dimension 0
Steps: 0%| | 0/1000 [00:02<?, ?it/s]
Traceback (most recent call last):
File "/usr/local/bin/accelerate", line 10, in
sys.exit(main())
^^^^^^
File "/usr/local/lib/python3.12/site-packages/accelerate/commands/accelerate_cli.py", line 50, in main
args.func(args)
File "/usr/local/lib/python3.12/site-packages/accelerate/commands/launch.py", line 1235, in launch_command
simple_launcher(args)
File "/usr/local/lib/python3.12/site-packages/accelerate/commands/launch.py", line 823, in simple_launcher
raise subprocess.CalledProcessError(returncode=process.returncode, cmd=cmd)
subprocess.CalledProcessError: Command '['/usr/local/bin/python', '/root/diffusers/examples/advanced_diffusion_training/train_dreambooth_lora_flux_advanced.py', '--mixed_precision=bf16', '--pretrained_model_name_or_path=black-forest-labs/FLUX.1-dev', '--dataset_name=Buattensorart2/r41s4', '--output_dir=/root/output_lora', '--instance_prompt=a photo of a r4is4 woman', '--image_column=image', '--caption_column=caption', '--class_prompt=a photo of a woman', '--class_data_dir=/root/class_imgs', '--resolution=1024', '--train_batch_size=1', '--with_prior_preservation', '--prior_loss_weight=1.0', '--optimizer=prodigy', '--gradient_accumulation_steps=1', '--learning_rate=1', '--lr_scheduler=constant', '--rank=32', '--lr_warmup_steps=0', '--max_train_steps=1000', '--checkpointing_steps=200', '--seed=230112', '--allow_tf32', '--gradient_checkpointing', '--cache_latents', '--train_batch_size=2', '--prior_generation_precision=bf16']' returned non-zero exit status 1.