segmentation fault

#24
by pangpangchao - opened

Hi, I download the weights. And install diffusers, and some related packages.
When I run the script

import torch
from diffusers import AutoPipelineForText2Image

pipeline = AutoPipelineForText2Image.from_pretrained(
    'path/to/my/donwload/OpenDalleV1.1', torch_dtype=torch.float16
).to('cuda')
image = pipeline(
    'black fluffy gorgeous dangerous cat animal creature, large orange eyes, big fluffy ears, piercing gaze, full moon, dark ambiance, best quality, extremely detailed'
).images[0]

, got the error message:

Loading pipeline components...: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 7/7 [00:01<00:00,  3.91it/s]
  0%|                                                                                                                                                                | 0/50 [00:00<?, ?it/s]
[1]    90516 segmentation fault (core dumped)  python test.py

The packages I used are as follows:

accelerate                0.25.0
diffusers                 0.25.0
torch                     2.1.2
tokenizers                0.15.0
transformers              4.36.2

Please, anyone can help me to figure out this issue? I saw the diffusers version in the repo is "_diffusers_version": "0.22.0.dev0".....
I am not sure what's the problem?

already fixed. caused my PyTorch version.... closed the discussion.

pangpangchao changed discussion status to closed

Sign up or log in to comment