Yuning You

Dear authors, Many thanks for your great efforts in implementation. We would like to sell our new algorithm JOAO https://arxiv.org/abs/2106.07594 (icml'21), with the purpose of automating graphcl (augmentation selection), to...

Performance on train from scratch

1

Thank you for your great work. I wonder what is the performance on train from scratch for cifar10 & cifar100? Seems this model heavily relies on the pretrained weights.

Failed to install pandas==0.18.0 in python 3.6.7

1

Hi Yu, Thanks for your excellent work. I met some issue while config the environment. Please see belows: 1. I failed to install pandas==0.18.0 in python 3.6.7, which output: ```...

Is there an implementation in huggingface?

1

Thanks for your great work. My question is whether there is a corresponding implementation in huggingface? I find this https://huggingface.co/armheb/DNA_bert_6?text=Paris+is+the+%5BMASK%5D+of+France. but not sure whether it is your official one.

Thanks for the nice collection effort. We recently published an ICLR'23 paper [Graph Domain Adaptation via Theory-Grounded Spectral Regularization](https://openreview.net/forum?id=OysfLgrk8mk) which provides a risk-bound analysis of the graph OOD problem. I...

sciplex3 count matrix gene indices

3

Thank you for the great resource. I notice that for sciplex3 data (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSM4150378) - The count matrix (GSM4150378_sciPlex3_A549_MCF7_K562_screen_UMI.count.matrix.gz) contains gene indices ranged in [1, **110981**]; - The gene annotations (GSM4150378_sciPlex3_A549_MCF7_K562_screen_gene.annotations.txt.gz)...

Request to add two graph self-supervised methods

6

## 🚀 Feature Request to add two graph self-supervised methods to Pytorch Geometric: (1) heuristic designed graph self-supervised tasks (ICML'20): https://arxiv.org/abs/2006.09136; code: https://github.com/Shen-Lab/SS-GCNs; (2) contrastive learning as self-supervision (NeurIPS'20): https://arxiv.org/abs/2010.13902;...