在reddit数据集上跑graphsage，loss异常

我用了examples目录下的graphsage，命令如下：
```
python run_graphsage.py --dataset reddit --run_mode train --batch_size 500 --learning_rate 0.01
```

运行结果如下：

请教一下各位，这个问题应该如何解决。十分感谢！

Start by reproducing the command in examples/run_graphsage.py with the reddit dataset and inspect the reported loss shown in the attached image. The issue does not identify the expected loss, failure point, or relevant test; done would require determining why the loss is abnormal and confirming corrected training behavior.

Written by the indexing model from the issue text.