blocked at 80%

[aeole75]

hello,
maybe i do something wrong but the train node stay stuck at 80%.. any idea ?

my database folder contains .png and .txt....
Do the pictures need to have all the same size ?

thank you !!!!

---

{
"sample_prompt": ""
}

train_func = run_lora_sd1_5
0%| | 0/10 [00:00<?, ?it/s]
100%|██████████| 10/10 [00:00<00:00, 384.63it/s]
INFO make buckets train_util.py:859
INFO number of images (including train_util.py:905
repeats) /
各bucketの画像枚数（繰り返
し回数を含む）
INFO bucket 0: resolution (512, 512), train_util.py:910
count: 150
INFO mean ar error (without repeats): train_util.py:915
0.0
INFO preparing accelerator train_network.py:225
INFO loading model for process 0/1 train_util.py:4385
accelerator device: cuda

_snapshot_download: runwayml/stable-diffusion-v1-5
Loading pipeline components...: 0%| | 0/6 [00:00<?, ?it/s]Some weights of the model checkpoint were not used when initializing CLIPTextModel:
['text_model.embeddings.position_ids']

Loading pipeline components...: 50%|█████ | 3/6 [00:00<00:00, 12.00it/s]
Loading pipeline components...: 83%|████████▎ | 5/6 [00:01<00:00, 2.71it/s]Some weights of the model checkpoint were not used when initializing AutoencoderKL:
['encoder.mid_block.attentions.0.to_to_k.bias, encoder.mid_block.attentions.0.to_to_k.weight, encoder.mid_block.attentions.0.to_to_q.bias, encoder.mid_block.attentions.0.to_to_q.weight, encoder.mid_block.attentions.0.to_to_v.bias, encoder.mid_block.attentions.0.to_to_v.weight, decoder.mid_block.attentions.0.to_to_k.bias, decoder.mid_block.attentions.0.to_to_k.weight, decoder.mid_block.attentions.0.to_to_q.bias, decoder.mid_block.attentions.0.to_to_q.weight, decoder.mid_block.attentions.0.to_to_v.bias, decoder.mid_block.attentions.0.to_to_v.weight']

Loading pipeline components...: 100%|██████████| 6/6 [00:01<00:00, 3.29it/s]
Loading pipeline components...: 100%|██████████| 6/6 [00:01<00:00, 3.52it/s]
You have disabled the safety checker for <class 'diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.StableDiffusionPipeline'> by passing safety_checker=None. Ensure that you abide to the conditions of the Stable Diffusion license and do not expose unfiltered results in services or applications open to the public. Both the diffusers team and Hugging Face strongly recommend to keep the safety filter enabled in all public facing circumstances, disabling it only for use-cases that involve analyzing network behavior or auditing its results. For more information, please have a look at huggingface/diffusers#254 .
2024-09-04 17:01:34 INFO UNet2DConditionModel: 64, 8, original_unet.py:1387
768, False, False
U-Net converted to original U-Net
import network module: networks.lora
Exception: Cannot copy out of meta tensor; no data! Please use torch.nn.Module.to_empty() instead of torch.nn.Module.to() when moving module from meta to a different device.