Problem using keyframe generation #223
Description
I forked your repo in hopes to use the awesome functionality of keyframing.
After I got it set up, I could do normal generation using only one image.
When I tried to use the keyframing feature in the sections I testet with 2 frames.
Directly after I started generation, I got the following error.
Is it common or did I do something wrong?
キーフレームの素晴らしい機能を使いたくて、あなたのリポジトリをフォークしました。
設定後、1枚の画像だけで通常の生成はできました。
テストしたセクションでキーフレーム機能を使おうと、2フレームで試したところ、
生成を開始した直後に以下のエラーが発生しました。
よくあることでしょうか？それとも何か間違った操作をしたのでしょうか？
An error occurred: 'numpy.ndarray' object has no attribute 'to' Traceback (most recent call last): File "E:\framepack_cu126_torch26\webui\endframe_ichi.py", line 1060, in worker start_latent = vae_encode(input_image_pt, vae) File "E:\framepack_cu126_torch26\system\python\lib\site-packages\torch\utils\_contextlib.py", line 116, in decorate_context return func(*args, **kwargs) File "E:\framepack_cu126_torch26\webui\diffusers_helper\hunyuan.py", line 109, in vae_encode latents = vae.encode(image.to(device=vae.device, dtype=vae.dtype)).latent_dist.sample() AttributeError: 'numpy.ndarray' object has no attribute 'to' Unloaded AutoencoderKLHunyuanVideo as complete. Unloaded SiglipVisionModel as complete. Unloaded AutoencoderKLHunyuanVideo as complete. Error occurred during cleanup: 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.