Partho Sarthi comments

Results 5 comments of


                                            Partho Sarthi

[FEA] Add UI support to the Profiling AutoTuner

Added a patch for UI for the AutoTuner. **Patch**: [Initial_Prototype_of_WebUI_for_AutoTuner.patch.zip](https://github.com/NVIDIA/spark-rapids/files/9330301/Initial_Prototype_of_WebUI_for_AutoTuner.patch.zip) **Sample WebUI:**

[bug] short flags/arguments does not work with `**kwargs`

Do we have any updates on this bug?

[AUDIT][SPARK][SQL] FileSourceScanExec supports push down data filter

Follow up for this work: Additional check was added to handle NPE. Spark Commit: https://github.com/apache/spark/commit/aa566ff3b9f

[AUDIT][SPARK-47164][SQL] Make Default Value From Wider Type Narrow Literal of v2 behave the same as v1

I thought this changed behaviour from V1 to V2 datasources for certain type. We should ensure we are consistent.

[BUG] parquet_test.py::test_buckets failed due to SparkOutOfMemoryError

This seems to reoccur in `rapids_it-UCX-shuffle-standalone/1246` job using `branch-25.04` ``` [2025-04-11T11:13:39.613Z] FAILED ../../src/main/python/parquet_test.py::test_buckets[-reader_confs11][DATAGEN_SEED=1744362284, TZ=UTC, IGNORE_ORDER, ALLOW_NON_GPU(DataWritingCommandExec,ExecutedCommandExec,WriteFilesExec)] - py4j.protocol.Py4JJavaError: An error occurred while calling o1117455.collectToPython. [2025-04-11T11:13:39.613Z] : org.apache.spark.SparkException: Job aborted due...