Anand created SPARK-53803:
-----------------------------
Summary: Add ArimaRegression for time series forecasting in MLlib
Key: SPARK-53803
URL: https://issues.apache.org/jira/browse/SPARK-53803
Project: Spark
Issue Type: New Feature
Components: ML, MLlib, PySpark
Affects Versions: 3.5.7
Reporter: Anand
The new components will implement the ARIMA (AutoRegressive Integrated Moving
Average) algorithm for univariate time series forecasting within the Spark ML
pipeline API.
This work will include:
- Implementation of ARIMA estimator with parameters (p, d, q)
- A fitted model `ArimaRegressionModel` for prediction
- Parameter support for (p, d, q) accessible from Scala and Python APIs
- PySpark bindings under `pyspark.ml.regression`
- Unit tests in Scala and Python for fit/transform, persistence, and predict
- An example usage added to `examples/ml/ArimaRegressionExample.scala`
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
