Hao Zhu
Hao Zhu
> @viadea since in some cases of XGBoost demo, we need to set spark.task.resource.gpu.amount=1 I agree. The ideal requirement is: By default, set spark.task.resource.gpu.amount to 1/spark.executor.cores. In Xgboost training/inference case,...
The real use case is like below: ``` from_utc_timestamp(cast(date_format(xxx, Some(Etc/UTC)) as timestamp), UTC) ``` So above is just my minimum reproduce. I can ask about the second parameter which is...
@res-life can you keep me updated for this feature once it is done? Thx.
Hi Team, I can reproduce the stackoverflow issue with ONLY AQE=on (Nothing to do with delta write optimize). Here is my minimum reproduce based on databricks 8.2ML GPU + 21.10...
The use case needs the Array[Binary] writing support as well. If I need to open a separate feature request , please feel free to let me know.
The goal is to make Array[Binary] work as well.
Here is an Array of Binary repro as well: ``` spark.sql("""CREATE TABLE spark_catalog.default.demo3 USING iceberg as select *, array(bd) as ab from spark_catalog.default.demo2""") spark.sql("""select * from spark_catalog.default.demo3""").collect ``` ``` !Exec...
@medb @mengdong @sameerz FYI
Note: 1. Using init script with Debian10: works fine. 2. Using custom image with Ubuntu: works fine. So this issue is specific to : Debian10 + custom image.
I tried but could not reproduce in house using my own STS. will try to get more information from user.