search

jefferyZhan / Griffon

【ECCV2024】The official repo of Griffon series
Apache License 2.0
93 stars 5 forks source link

RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cuda:1 and cuda:0 #10

Closed dawn-ech closed 4 months ago

dawn-ech commented 4 months ago

Thanks for your great work! I am trying to run demo.sh on 2x 3090, but I get this error. The problem is located in llava/model/multimodal_encoder.py image_forward_outs = self.vision_tower(images, output_hidden_states=True). Here, images is on cuda:0, but I print its device in forward function (transformers.models.clip.CLIPVisionModel.forward), getting cuda:1. Could you give me some suggestions about it?

jefferyZhan commented 4 months ago

Hi @dawn-ech, This is caused by the auto setting in the weight loading code following llava. Add CUDA_VISIBLE_DEVICES setting will solve this, like: CUDA_VISIBLE_DEVICES=0 bash demo/demo.sh demo/1vn.jpg "In this picture, identify and locate all the people in the front."

dawn-ech commented 4 months ago

This problem is solved. Thanks!