StevenLiuWen

hi @wiamadaya , it seems that the buffer is not full, **Filling up shuffle buffer (this may take a while): 500 of 1000**, and there might not be enough space...

@1zb , Thanks for sharing the preprocessed data and releasing the code. It seems that google drive has limited the download yet, and currently we can not download the preprocessed...

Hi, @OldChi , You need to set the flag `--image_size 512` to generate 512 x 512 resolution images when you have trained the model on such resolution.

@cwzat we update the additional files of iPER https://onedrive.live.com/?authkey=%21AJL_NAQMkdXGPlA&id=3705E349C336415F%2188052&cid=3705E349C336415F, and the training details are shown in [https://github.com/svip-lab/impersonator/blob/master/doc/train.md](url). You can follow the training script to train the iPER from scratch.

@Tayfur26 @haviduck we update the additional files of iPER https://onedrive.live.com/?authkey=%21AJL_NAQMkdXGPlA&id=3705E349C336415F%2188052&cid=3705E349C336415F, and the training details are shown in [https://github.com/svip-lab/impersonator/blob/master/doc/train.md](url). Welcome to clone the latest codes, and follow the training script to...

I have encountered the same issue in CentOS 7, GCC-7.5 (and both GCC 7.3), and Cuda 10.2.

Hi, the model supports multiple images. Refer to https://huggingface.co/deepseek-ai/deepseek-vl-7b-chat/discussions/4 for the prompt example. ```python conversation = [ { "role": "User", "content": "Compare and contrast and .", "images": ["./data/1.png", "./data/2.jpg"] },...

It could be this one https://arxiv.org/pdf/2405.20324 (Nicolas Dufour et. al, CVPR 2024) which has extended the RIN into text condition.

> @StevenLiuWen very cool! and not the original author(s)! Also, another work, PointInfinity (https://arxiv.org/pdf/2404.03566) applied it to the 3D point cloud generation. RIN or perceiver-io style architecture has a nice...