topic Re: SparkException: Job aborted due to stage failure when attempting to run grid_pointascellid in Get Started Discussions

SparkException: Job aborted due to stage failure when attempting to run grid_pointascellid

kll — Fri, 07 Jul 2023 14:23:28 GMT

I am attempting to apply Mosaic's `grid_pointascellid` method on a spark dataframe with `lat`, `lon` columns.

```
import pyspark.sql.functions as F

# Create a Spark DataFrame with a lat and lon column
df = spark.createDataFrame([
("point1", 10.0, 20.0),
("point2", 30.0, 40.0),
("point3", 50.0, 60.0),
], ["name", "lat", "lon"])

# Print the results of the DataFrame
df.show()

df1 = df.withColumn("id", grid_pointascellid(st_point(df["lon"], df["lat"]), lit(7)))
df1.show()

```

The schema of the dataframe is:

```

|-- name: string (nullable = true)
|-- lat: decimal(9,6) (nullable = true)
|-- lng: decimal(9,6) (nullable = true)

```

Exception:

org.apache.spark.SparkException: Job aborted due to stage failure: Task 0 in stage 251.0 failed 4 times, most recent failure: Lost task 0.3 in stage 251.0 (TID 15016) (10.138.221.80 executor 50): java.lang.ClassCastException: org.apache.spark.sql.types.Decimal cannot be cast to java.lang.Double

https://databrickslabs.github.io/mosaic/api/spatial-indexing.html?highlight=grid_boundary#grid-pointascellid

Re: SparkException: Job aborted due to stage failure when attempting to run grid_pointascellid

Lakshay — Fri, 07 Jul 2023 14:16:00 GMT

At which step are you getting the error? Also, can you share the code for "grid_pointascellid" function to understand what it is doing?

Re: SparkException: Job aborted due to stage failure when attempting to run grid_pointascellid

kll — Fri, 07 Jul 2023 14:27:53 GMT

@Lakshay

On this step, when I attempt to apply the method:

```

df1 = df.withColumn("id", grid_pointascellid(st_point(df["lon"], df["lat"]), lit(7)))
df1.show()

```

Re: SparkException: Job aborted due to stage failure when attempting to run grid_pointascellid

Anonymous — Wed, 12 Jul 2023 09:42:53 GMT

Hi @kll

Hope you are well. Just wanted to see if you were able to find an answer to your question and would you like to mark an answer as best? It would be really helpful for the other members too.

Cheers!

Re: SparkException: Job aborted due to stage failure when attempting to run grid_pointascellid

Tharun-Kumar — Fri, 14 Jul 2023 19:02:40 GMT

@kll

This error appears because the function grid_pointascellid expects a Double type column and a Decimal column type was provided as input.

To overcome this, before you apply the grid_pointascellid, I would recommend casting the columns lat and lon.

df1 = df.withColumn("lat", df["lat"].cast("double")).withColumn("lon", df["lon"].cast("double"))

df2 = df1.withColumn("id", grid_pointascellid(st_point(df1["lon"], df1["lat"]), lit(7)))
df2.show()

By performing this explicit cast before calling the mosaic functions, i was able to get the output below.

| name. | lat. | lon. | id |
|point1. |10.0. |20.0. |609880166050562047|
|point2. |30.0 |40.0. |609448747927076863|
|point3. |50.0 |60.0. |608566547836829695|