Xianyang Liu
Xianyang Liu
We have already done this by leverage arrow format. You can see https://github.com/oap-project/raydp/blob/1f4853f44476fad52b9d2fa4519c4c76974557ef/python/raydp/spark/dataset.py#L70. This will save the spark dataframe into the ray object store and get a list of ray...
Just a suggestion. In production, the software is updated much more slowly than the upstream update/release. It would be better to support more versions of basic services in our product.
Not yet, you could use the pip under the conda env right now. We will add it later, but with low priority.
Sure, thanks for your interest. We should do it, but we have limited resources. It should be added later.
Hi @klwuibm, the feature is not supported yet. @kira-lin is working to support it.
Hi Zhe, it seems like the port 25333 is occupied. 发自我的iPhone 在 2021年2月17日，下午2:32，Zhe Zhang 写道： I was using very simple code: import ray import raydp ray.init() spark = raydp.init_spark(app_name="RayDP...
Hi @rdblue @kbendick @szehon-ho could you help to review this when you are free? Thanks a lot.
> so partition-spec-id on the header of manifestFile is always the same as partition-spec-id of the manifest-file entry From here: https://github.com/apache/iceberg/blob/master/api/src/main/java/org/apache/iceberg/ManifestFile.java#L34. I think it should be. And the partition spec...
Thanks @szehon-ho @rdblue @nastra @zinking
Hi @rdblue @kbendick, could you help to review this when you are free? Thanks a lot.