feat: compressor data

CHANGELOG.md

@@ -1,3 +1,9 @@
+## 0.1.0-rc.1 (2026-03-30)
+
+* feat: compressor data ([1685d7d](https://github.com/sequential-parameter-optimization/spotdesirability/commit/1685d7d))
+* ci: install spotdesirability jupyter kernel using --sys-prefix instead of --user ([d5139de](https://github.com/sequential-parameter-optimization/spotdesirability/commit/d5139de))
+* ci: remove explicit jupyter kernel dependencies to fix CI kernel resolution ([18402c5](https://github.com/sequential-parameter-optimization/spotdesirability/commit/18402c5))
+* Update uv.lock ([815d778](https://github.com/sequential-parameter-optimization/spotdesirability/commit/815d778))
 ## <small>0.0.27 (2026-02-28)</small>
 
 * Merge branch 'main' into develop ([1b23893](https://github.com/sequential-parameter-optimization/spotdesirability/commit/1b23893))

MANIFEST.in

@@ -0,0 +1 @@
+recursive-include src/spotdesirability/datasets *.csv

pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "uv_build"
 
 [project]
 name = "spotdesirability"
-version = "0.0.27"
+version = "0.1.0rc1"
 authors = [
   { name="T. Bartz-Beielstein", email="tbb@bartzundbartz.de" }
 ]

src/spotdesirability/data_utils.py

@@ -1,10 +1,51 @@
 import os
+import importlib.resources as pkg_resources
+from typing import Tuple
 
+import pandas as pd
 
-def get_data_folder_path():
+
+def get_data_folder_path() -> str:
     """Returns the absolute path to the data folder located in the package."""
     # Assume the 'data' directory is within the same package directory
     current_file_path = os.path.abspath(__file__)
     package_dir = os.path.dirname(current_file_path)
     data_folder_path = os.path.join(package_dir, "data")
     return data_folder_path
+
+
+def load_compressor_data() -> Tuple[pd.DataFrame, pd.DataFrame]:
+    """Loads the normalized compressor datasets X and Z.
+
+    This function reads df_x_normalized.csv and df_z_normalized.csv from the
+    spotdesirability.datasets module and returns them as pandas DataFrames.
+    The datasets represent the values of $X$ and $Z$ respectively.
+
+    Returns:
+        Tuple[pd.DataFrame, pd.DataFrame]: A tuple containing the datasets X and Z
+        as pandas DataFrames respectively.
+
+    Raises:
+        FileNotFoundError: If the datasets cannot be found.
+        RuntimeError: If there is an issue reading the CSV data.
+
+    Example:
+        ```python
+        import spotdesirability.data_utils as du
+
+        df_x, df_z = du.load_compressor_data()
+        ```
+    """
+    try:
+        x_path = pkg_resources.files("spotdesirability.datasets") / "df_x_normalized.csv"
+        z_path = pkg_resources.files("spotdesirability.datasets") / "df_z_normalized.csv"
+
+        with pkg_resources.as_file(x_path) as p_x, pkg_resources.as_file(z_path) as p_z:
+            df_x = pd.read_csv(p_x)
+            df_z = pd.read_csv(p_z)
+
+        return df_x, df_z
+    except FileNotFoundError as fnf_err:
+        raise fnf_err
+    except Exception as err:
+        raise RuntimeError(f"Error loading compressor datasets: {err}") from err

tests/test_data_utils.py

@@ -0,0 +1,16 @@
+import pandas as pd
+import spotdesirability.data_utils as du
+
+def test_load_compressor_data():
+    """Test if load_compressor_data returns two pandas DataFrames with expected data."""
+    df_x, df_z = du.load_compressor_data()
+
+    assert isinstance(df_x, pd.DataFrame), "Expected df_x to be a pandas DataFrame."
+    assert isinstance(df_z, pd.DataFrame), "Expected df_z to be a pandas DataFrame."
+
+    assert not df_x.empty, "df_x DataFrame is empty."
+    assert not df_z.empty, "df_z DataFrame is empty."
+
+    # Check that they have a reasonable number of columns or rows, ensuring successful parse
+    assert len(df_x.columns) > 0, "df_x has no columns."
+    assert len(df_z.columns) > 0, "df_z has no columns."