added sample slides for demo presentation and added trainscript for demo model

Signed-off-by: Christoph Huy <christoph.huy@campus.tu-berlin.de>

Author: Perimora

amos_team_resources/demoday/amos_demo.pptx

@@ -0,0 +1,57 @@
+import sys
+
+from pathlib import Path
+from pyspark.sql import SparkSession
+
+DATA = Path(__file__).parent.parent / "data"
+
+print(f"Data directory is located at: {DATA}")
+
+
+def setup_python_env():
+    project_root = Path(__file__).parent.parent.parent.parent
+    print(f"Project root directory is located at: {project_root}")
+
+    sdk_path = project_root / "src" / "sdk" / "python"
+    sdk_path = sdk_path.resolve()
+
+    sys.path.insert(0, str(sdk_path))
+
+
+def main():
+
+    print("Setting up Python environment...")
+    setup_python_env()
+    from rtdip_sdk.pipelines.forecasting.spark.autogluon_timeseries import (
+        AutoGluonTimeSeries,
+    )
+
+    print("Starting Spark session...")
+    spark = (
+        SparkSession.builder.master("local[*]")
+        .appName("SCADA-Forecasting")
+        .config("spark.driver.memory", "8g")
+        .config("spark.executor.memory", "8g")
+        .config("spark.driver.maxResultSize", "2g")
+        .config("spark.sql.shuffle.partitions", "50")
+        .config("spark.sql.execution.arrow.pyspark.enabled", "true")
+        .getOrCreate()
+    )
+
+    print("Reading preprocessed SCADA data...")
+
+    data_path = DATA / "scada_prepro.parquet"
+    assert data_path.exists(), f"Data file not found at {data_path}"
+    df = spark.read.parquet(str(data_path))
+
+    print("Starting AutoGluon Training...")
+    ag_model = AutoGluonTimeSeries()
+    train_df, test_df = ag_model.split_data(df)
+    res_dict = ag_model.train(train_df)
+
+    print("Saving test dataset...")
+    test_df.write.mode("overwrite").parquet(str(DATA / "scada_test.parquet"))
+
+
+if __name__ == "__main__":
+    exit(main())