AI/diffusers
1
0
Fork 0
mirror of https://github.com/huggingface/diffusers.git synced 2026-01-27 17:22:53 +03:00
Code Activity

[Qwen LoRA training] fix bug when offloading (#12440)

Browse Source 
* fix bug when offload and cache_latents both enabled

* fix bug when offload and cache_latents both enabled

* fix bug when offload and cache_latents both enabled

* fix bug when offload and cache_latents both enabled

* fix bug when offload and cache_latents both enabled

* fix bug when offload and cache_latents both enabled

* fix bug when offload and cache_latents both enabled

* fix bug when offload and cache_latents both enabled

* fix bug when offload and cache_latents both enabled
This commit is contained in:
Linoy Tsaban
2025-10-07 18:27:15 +03:00
committed by GitHub
parent 2d69bacb00
commit 1066de8c69
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
examples/dreambooth/train_dreambooth_lora_qwen_image.py
View File

@@ -1338,7 +1338,7 @@ def main(args):
batch["pixel_values"] = batch["pixel_values"].to(
accelerator.device, non_blocking=True, dtype=vae.dtype
)
latents_cache.append(vae.encode(batch["pixel_values"]).latent_dist)
latents_cache.append(vae.encode(batch["pixel_values"]).latent_dist)
if train_dataset.custom_instance_prompts:
with offload_models(text_encoding_pipeline, device=accelerator.device, offload=args.offload):
prompt_embeds, prompt_embeds_mask = compute_text_embeddings(