search

sherwinbahmani / 4dfy

4D-fy: Text-to-4D Generation Using Hybrid Score Distillation Sampling
https://sherwinbahmani.github.io/4dfy/
Apache License 2.0
309 stars 8 forks source link

Version Problem #7

Closed zimingzhong closed 9 months ago

zimingzhong commented 9 months ago

we have to use the low version 'xformers' like 0.0.12 to satisfy torch==1.12.1+cu113. If using high version xformers, it will automatically download torch=2.1.2.

but low xformers.ops.memory_efficient_attention get the version problem and i have no idea to fix it.


4dfy/extern/MVDream/mvdream/ldm/modules/diffusionmodules/model.py", line 258, in forward

 out = xformers.ops.memory_efficient_attention(q, k, v, attn_bias=None, op=self.attention_op)

TypeError: memory_efficient_attention() got an unexpected keyword argument 'attn_bias'
sherwinbahmani commented 9 months ago

Hi,

I am using torch==2.0.1 and xformers==0.0.20

Does that work for you?

zimingzhong commented 9 months ago

Hi,

I am using torch==2.0.1 and xformers==0.0.20

Does that work for you?

thanks for your info. i want to know what the version of xformers when the torch==1.12.1+cu113. And i am trying reinstall torch=2.0.1.

sherwinbahmani commented 9 months ago

I have not tried it with torch 1.12.1. If you can, try torch 2.0.1. Also see in the original threestudio github if there were problems with 1.12.1, if you want to stick with that

zimingzhong commented 9 months ago

I have not tried it with torch 1.12.1. If you can, try torch 2.0.1. Also see in the original threestudio github if there were problems with 1.12.1, if you want to stick with that

thanks for you quick reply. can you show the version of your torch torchvision cuda and xformers?

sherwinbahmani commented 9 months ago

torch==2.0.1+cu117 torchvision==0.15.2+cu117 xformers==0.0.20

zimingzhong commented 9 months ago

torch==2.0.1+cu117 torchvision==0.15.2+cu117 xformers==0.0.20

thank you!