drprojects / superpoint_transformer

Official PyTorch implementation of Superpoint Transformer introduced in [ICCV'23] "Efficient 3D Semantic Segmentation with Superpoint Transformer" and SuperCluster introduced in [3DV'24 Oral] "Scalable 3D Panoptic Segmentation As Superpoint Graph Clustering"
MIT License
560 stars 72 forks source link

Fix CUDA version in install.sh to be coherent with pyg wheels (cu 11.8) #40

Closed CharlesGaydon closed 10 months ago

CharlesGaydon commented 10 months ago

Still need to test locally, hence the draft PR.

What does this PR do?

This PR sets the index url when pip installin torch. This was necessary when installing the env in Google Collab, since torch would be installed with cuda 12.0 while pytorch geometric would be later installed wit cuda 11.8, resulting in an error message.

Before submitting

Did you have fun?

Make sure you had fun coding 🙃

drprojects commented 10 months ago

Hi @CharlesGaydon thanks for the PR. Seems reasonable ! Can I let you do the testing before I validate this ?

CharlesGaydon commented 10 months ago

Sure thing. Just wanted to open it to have it ready somewhere, but it is not done yet.

CharlesGaydon commented 10 months ago

🥳