PKU-DAIR / SGL

A scalable graph learning toolkit for extremely large graph datasets. (WWW'22, 🏆 Best Student Paper Award)
https://sgl-doc.readthedocs.io/en/latest
MIT License
140 stars 21 forks source link

Where is PASCA-V1 (V2, V3) #2

Closed talentedx closed 2 years ago

talentedx commented 2 years ago

Hi, Your paper is really interesting, and I try to retype this code and find that PASCA-V1(V2, V3) are not included in this respority. From this code, we know that SGAP is a fundamental framework of this paper, and SGC is constructed based on this framework, am I right?

zwt233 commented 2 years ago

From this code, we know that SGAP is a fundamental framework of this paper, and SGC is constructed based on this framework, am I right?

Yes, we adopt the scalable design paradigm (SGAP) in our GNN architecture design. Based on SGAP, we have implemented many scalable GNNs (including SGC) in https://github.com/PKU-DAIR/SGL/tree/main/sgl/models. Besides, PASCA-V1(V2, V3) are searched by our system with the multiple objectives (i.e., predictive accuracy and inference time), and we will add the support of these three scalable models ASAP.

talentedx commented 2 years ago

From this code, we know that SGAP is a fundamental framework of this paper, and SGC is constructed based on this framework, am I right?

Yes, we adopt the scalable design paradigm (SGAP) in our GNN architecture design. Based on SGAP, we have implemented many scalable GNNs (including SGC) in https://github.com/PKU-DAIR/SGL/tree/main/sgl/models. Besides, PASCA-V1(V2, V3) are searched by our system with the multiple objectives (i.e., predictive accuracy and inference time), and we will add the support of these three scalable models ASAP.

OK, I am looking forward to the updated SGL including PASCA-V1, V2, V3.

zwt233 commented 2 years ago

From this code, we know that SGAP is a fundamental framework of this paper, and SGC is constructed based on this framework, am I right?

Yes, we adopt the scalable design paradigm (SGAP) in our GNN architecture design. Based on SGAP, we have implemented many scalable GNNs (including SGC) in https://github.com/PKU-DAIR/SGL/tree/main/sgl/models. Besides, PASCA-V1(V2, V3) are searched by our system with the multiple objectives (i.e., predictive accuracy and inference time), and we will add the support of these three scalable models ASAP.

OK, I am looking forward to the updated SGL including PASCA-V1, V2, V3.

Thanks for your attention! Implementing scalable GNNs with SGL is quite simple with our pre-defined operators, and we will continuously add more scalable GNN models for both homogeneous and heterogeneous graphs.