Drew

[Question]How to store additional feature in GraphScope

2

您好，咨询下GraphScope中怎么存储多余的一些字段： 例如，在看vectornet实现代码时，它的每个子图包含额外的字段cluster, valid_len, 其中一条数据的shape可以如下 > 该代码是在PYG基础上实现的 > https://github.com/xk-huang/yet-another-vectornet/blob/master/modeling/vectornet.py#L49 ```Python data (Data): [x, y, cluster, edge_index, valid_len] # 一条数据的shape Data(cluster[253], edge_index=[2, 2342], time_step_len=[1], valid_len=[1], x=[253, 8], y=[60]) ``` - `cluster`, `time_step_len`,...

add coverage compile option

1

add coverage compile option for better code testing rate detect

There is no Compile option to run converage test for abseil

1

**Describe the bug** - here in abseil-cpp repo, we unable to know the converege rate, but many of other repo have this - this is import for us to know...

- 看着官网只给了hologres本身的读写性能测试结果 https://help.aliyun.com/document_detail/605017.html - Flink/spark对接hologres，看着官方提到走Connector写入这种方式，LBS --> private api service--> backend (在插入数据场景，这里存在两次数据拷贝吧？），Flink对接hologres的写数据性能有评测么？**请教一下性能怎么样**，上面两次拷贝是否可以优化 https://developer.aliyun.com/article/778798