zhengruifeng opened a new pull request, #58624: URL: https://github.com/apache/spark/pull/58624
### What changes were proposed in this pull request? This PR removes the redundant `GeneralizedLinearRegressionModel.transformImpl` override and keeps its implementation directly in `transform`. ### Why are the changes needed? `GeneralizedLinearRegressionModel` must override `transform` because it can produce `linkPredictionCol` when `predictionCol` is empty, while the base `Predictor` implementation only invokes `transformImpl` when `predictionCol` is set. Once GLR owns `transform`, the separate `transformImpl` override is only called by that method and adds an unnecessary second `transformSchema` invocation. Inlining the implementation removes this redundant indirection and schema validation without changing the generated columns. ### Does this PR introduce _any_ user-facing change? No. ### How was this patch tested? The following checks passed: ``` build/sbt mllib/compile build/sbt 'mllib/testOnly org.apache.spark.ml.regression.GeneralizedLinearRegressionSuite' ``` `GeneralizedLinearRegressionSuite` ran 29 tests successfully, with one ignored test. ### Was this patch authored or co-authored using generative AI tooling? Generated-by: OpenAI Codex (GPT-5) -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
