CogDL

CogDL copied to clipboard

RuntimeError: CUDA out of memory with 6M nodes, 8M edges on A100 GPU

1

## 🐛 Bug ``` |-------------------------------------------------------------------------------------------------------| *** Running (`tmp_data.pt`, `unsup_graphsage`, `node_classification_dw`, `unsup_graphsage_mw`) |-------------------------------------------------------------------------------------------------------| Model Parameters: 1568 0%| | 0/500 [00:00

chi2liu

## Description update experiment & trainer and add examples/ogb/molhiv - update experiment - update trainer - create a example in examples/ogb/molhiv

KunZhang99

Questions about big dataset processing or split reading

7

## ❓ Questions & Help Hi, I am a beginer in gnn, thanks for the convenient tool COGDL open sourced by THUDM. ## Question Description I implemented a customized Dataset...

odymit

Embedding学习任务并行化

1

## ❓ Questions & Help 观察到cogdl中Embedding学习任务似乎没有提供并行化的工具，后续会有改进吗

nanyoullm

请问如何更改评价标准

1

我想把HAN模型的ACC改成AP，请问可以在哪里改

csyifan

Could Cogdl implement graph regression task？

2

I used MLP as the regression layer,but got an error that _ValueError: Expected 2 or more dimensions (got 1)_ Is that couldn't implement graph regression task,please?

huang429

最新版的cogdl如何获取有监督节点分类的结果

8

非常感谢你们开发并开源cogdl，请问怎样获得有监督节点分类的结果而不是只显示acc呢？

huihuijiangqiang

提交数据集

2

您好，我想提交一个数据集

YuanSQ1997

pipeline使用无监督GNN模型出错

6

## ❓ Questions & Help 您好，我在使用cogdl中的pipelin函数进行图嵌入的时候遇到了问题。当我使用以“prone”为首的emb_models时，一切正常，而当我使用以“dgi”为首的gnn_models时，出现了一些问题。 代码如下： import numpy as np import torch from cogdl import pipeline # generate embedding by an unweighted graph edge_index = np.array([[0, 1], [0, 2],...

Saberfish

Customized graph_classification dataset & graph2vec

2

## 🐛 Bug CogDL requires the features of the a custom data set to be a tensor, which can not contain strings, and wl_iterations uses [str.join](https://docs.python.org/3/library/stdtypes.html#str.join), which errors if there...

Cedric-Kram