[WIP] Almost match Cast Floats to Decimal #10909
Closed
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Signed-off-by: Haoyang Li <haoyangl@nvidia.com>
Signed-off-by: Haoyang Li <haoyangl@nvidia.com>
|
cc @ttnghia Here are some integration tests. scala test and Spark UT are also useful. # Integration test:
./integration_tests/run_pyspark_from_build.sh -s -k test_cast_float_to_decimal
# Spark UT:
mvn test -Dbuildver=330 -DwildcardSuites=org.apache.spark.sql.rapids.suites.RapidsCastSuite
# Scala Test:
mvn test -Dbuildver=330 -DwildcardSuites=com.nvidia.spark.rapids.CastOpSuite |
thirtiseven
commented
May 27, 2024
| @@ -877,8 +877,8 @@ class CastOpSuite extends GpuExpressionTestSuite { | |||
|
|
|||
| overflowCase(DataTypes.FloatType, precision = 10, scale = 6, | |||
| generator = floatGenerator(Seq(12345.678f))) | |||
| overflowCase(DataTypes.DoubleType, precision = 15, scale = -5, | |||
| generator = doubleGenerator(Seq(1.23e21))) | |||
| // overflowCase(DataTypes.DoubleType, precision = 15, scale = -5, | |||
There was a problem hiding this comment.
These two line failed because #10908
| @@ -261,6 +261,23 @@ def test_cast_long_to_decimal_overflow(): | |||
| lambda spark : unary_op_df(spark, long_gen).select( | |||
| f.col('a').cast(DecimalType(18, -1)))) | |||
|
|
|||
| @datagen_overrides(seed=0, reason='edge cases') | |||
There was a problem hiding this comment.
Fixed seed because #10890
|
Close because we have a better approach in #10917 |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Post this pr for sharing tests.
Fixes #9682
Fixes #10809
This PR rewrite Cast floats to decimal with a new floats => string => decimal path, to make the results closer to CPU.
There are still some differences from the known limits of ryu float to string and two edge cases in string to decimal #10890 and #10908.
Performance test to cast 5000000 floats to 10 kinds of decimal types (in ms)
We can make it a bit faster by combining the two kernels in jni.