The introduction of the ruturn of libcp.cutpursuit
shaofeifei11 opened this issue · comments
I am sorry, I can not understand what is the "components", "in_component" (lines 177 of partition/partition.py )
components, in_component = libcp.cutpursuit(features, graph_nn["source"], graph_nn["target"] , graph_nn["edge_weight"], args.reg_strength)
In particular, "in" in "in_component" is an abbreviation of "index"?
Could you briefly introduce what data is stored in "component" and "In_component"?
I am very expected to get your help. Thank you!
hello, i have the same question, do you understand them now?
I think that in_component is the number of components which this point is belonging to.
hello, i have the same question, do you understand them now?
Hi, sorry for delayed answer.
components is a list of size the number of components. components[c] is the list of the indices of the vertices in component c.
in_component is an array of size n_vertex. in_component[v] returns the index of the component containing v.
In other words, in_component[components[c]]==c
Hi!
We are releasing a new version of SuperPoint Graph called SuperPoint Transformer (SPT).
It is better in any way:
| ✨ SPT in numbers ✨ |
|---|
| 📊 SOTA results: 76.0 mIoU S3DIS 6-Fold, 63.5 mIoU on KITTI-360 Val, 79.6 mIoU on DALES |
| 🦋 212k parameters only! |
| ⚡ Trains on S3DIS in 3h on 1 GPU |
| ⚡ Preprocessing is x7 faster than SPG! |
| 🚀 Easy install (no more boost!) |
If you are interested in lightweight, high-performance 3D deep learning, you should check it out. In the meantime, we will finally retire SPG and stop maintaining this repo.