Use actual Enum for constants and types

src/paradigma/constants.py

@@ -1,4 +1,6 @@
-class DataColumns:
+from enum import Enum
+
+class DataColumns(Enum):
     """
     Enum for the data channels in tsdf.
     """
@@ -11,7 +13,7 @@ class DataColumns:
     PPG = "green"
     TIME = "time"
 
-class DataUnits:
+class DataUnits(Enum):
     """
     Enum for the data channel unit types in tsdf.
     """
@@ -21,7 +23,7 @@ class DataUnits:
     """ The rotation is in degrees per second. """
 
 
-class TimeUnit:
+class TimeUnit(Enum):
     """
     Enum for the `time` channel unit types in tsdf.
     """

src/paradigma/gait_analysis.py

@@ -122,9 +122,9 @@ def extract_arm_swing_features(input_path: Union[str, Path], output_path: Union[
     # perform principal component analysis on the gyroscope signals to obtain the angular velocity in the
     # direction of the swing of the arm 
     df[config.velocity_colname] = pca_transform_gyroscope(
-        df=df, 
-        y_gyro_colname=DataColumns.GYROSCOPE_Y,
-        z_gyro_colname=DataColumns.GYROSCOPE_Z,
+        df=df,
+        y_gyro_colname=DataColumns.GYROSCOPE_Y.value,
+        z_gyro_colname=DataColumns.GYROSCOPE_Z.value,
         pred_gait_colname=config.pred_gait_colname
     )
 

src/paradigma/gait_analysis_config.py

@@ -30,10 +30,10 @@ def __init__(self) -> None:
 
         self.time_colname = 'time'
 
-        self.l_accelerometer_cols: List[str] = [
-            DataColumns.ACCELEROMETER_X, 
-            DataColumns.ACCELEROMETER_Y, 
-            DataColumns.ACCELEROMETER_Z
+        self.l_accelerometer_cols: List[str] = [ # strings, not DataColumns, as we'll edit them
+            "accelerometer_x",
+            "accelerometer_y",
+            "accelerometer_z",
         ]
 
         self.l_gravity_cols: List[str] = [f'grav_{x}' for x in self.l_accelerometer_cols]
@@ -95,7 +95,7 @@ def __init__(self) -> None:
         self.time_filename = 'gait_time.bin'
         self.values_filename = 'gait_values.bin'
 
-        self.l_accel_cols = [DataColumns.ACCELEROMETER_X, DataColumns.ACCELEROMETER_Y, DataColumns.ACCELEROMETER_Z]
+        self.l_accel_cols = [DataColumns.ACCELEROMETER_X.value, DataColumns.ACCELEROMETER_Y.value, DataColumns.ACCELEROMETER_Z.value]
 
         self.time_colname = 'time'
 
@@ -140,15 +140,15 @@ def initialize_column_names(self, time_colname='time', pred_gait_colname='pred_g
         self.segment_nr_colname = segment_nr_colname
 
         self.l_accelerometer_cols: List[str] = [
-            DataColumns.ACCELEROMETER_X, 
-            DataColumns.ACCELEROMETER_Y, 
-            DataColumns.ACCELEROMETER_Z
+            DataColumns.ACCELEROMETER_X.value, 
+            DataColumns.ACCELEROMETER_Y.value, 
+            DataColumns.ACCELEROMETER_Z.value
         ]
 
         self.l_gyroscope_cols: List[str] = [
-            DataColumns.GYROSCOPE_X, 
-            DataColumns.GYROSCOPE_Y, 
-            DataColumns.GYROSCOPE_Z
+            DataColumns.GYROSCOPE_X.value, 
+            DataColumns.GYROSCOPE_Y.value, 
+            DataColumns.GYROSCOPE_Z.value
         ]
 
         self.l_gravity_cols: List[str] = [f'grav_{x}' for x in self.l_accelerometer_cols]
@@ -225,8 +225,12 @@ def __init__(self) -> None:
         self.time_filename = 'arm_swing_time.bin'
         self.values_filename = 'arm_swing_values.bin'
 
-        self.l_accel_cols = [DataColumns.ACCELEROMETER_X, DataColumns.ACCELEROMETER_Y, DataColumns.ACCELEROMETER_Z]
-        self.l_gyro_cols = [DataColumns.GYROSCOPE_X, DataColumns.GYROSCOPE_Y, DataColumns.GYROSCOPE_Z]
+        self.l_accel_cols = [DataColumns.ACCELEROMETER_X.value,
+                             DataColumns.ACCELEROMETER_Y.value,
+                             DataColumns.ACCELEROMETER_Z.value]
+        self.l_gyro_cols = [DataColumns.GYROSCOPE_X.value,
+                            DataColumns.GYROSCOPE_Y.value,
+                            DataColumns.GYROSCOPE_Z.value]
 
 
 class ArmSwingQuantificationConfig:

src/paradigma/heart_rate_analysis.py

@@ -16,12 +16,12 @@ def extract_signal_quality_features(input_path: str, classifier_path: str, outpu
     # load data
     metadata_time_ppg, metadata_samples_ppg = read_metadata(input_path, "PPG_meta.json", "PPG_time.bin", "PPG_samples.bin")
     df_ppg = tsdf.load_dataframe_from_binaries([metadata_time_ppg, metadata_samples_ppg], tsdf.constants.ConcatenationType.columns)
-    arr_ppg = df_ppg[DataColumns.PPG].to_numpy()
-    relative_time_ppg = df_ppg[DataColumns.TIME].to_numpy()
+    arr_ppg = df_ppg[DataColumns.PPG.value].to_numpy()
+    relative_time_ppg = df_ppg[DataColumns.TIME.value].to_numpy()
 
     metadata_time_acc, metadata_samples_acc = read_metadata(input_path, "accelerometer_meta.json", "accelerometer_time.bin", "accelerometer_samples.bin")
     df_acc = tsdf.load_dataframe_from_binaries([metadata_time_acc, metadata_samples_acc], tsdf.constants.ConcatenationType.columns)
-    arr_acc = df_acc[[DataColumns.ACCELEROMETER_X, DataColumns.ACCELEROMETER_Y, DataColumns.ACCELEROMETER_Z]].to_numpy()
+    arr_acc = df_acc[[DataColumns.ACCELEROMETER_X.value, DataColumns.ACCELEROMETER_Y.value, DataColumns.ACCELEROMETER_Z.value]].to_numpy()
 
     sampling_frequency_ppg = config.sampling_frequency_ppg
     sampling_frequency_imu = config.sampling_frequency_imu

src/paradigma/imu_preprocessing.py

@@ -24,7 +24,7 @@ def preprocess_imu_data(input_path: Union[str, Path], output_path: Union[str, Pa
 
     # convert to relative seconds from delta milliseconds
     df[config.time_colname] = transform_time_array(
-        time_array=df[config.time_colname],
+        time_array=df[config.time_colname.value],
         scale_factor=1000, 
         input_unit_type = TimeUnit.difference_ms,
         output_unit_type = TimeUnit.relative_ms)
@@ -39,7 +39,7 @@ def preprocess_imu_data(input_path: Union[str, Path], output_path: Union[str, Pa
         resampling_frequency=config.sampling_frequency)
 
     if config.side_watch == 'left':
-        df[DataColumns.ACCELEROMETER_X] *= -1
+        df[DataColumns.ACCELEROMETER_X.value] *= -1
 
     for col in config.d_channels_accelerometer.keys():
 
@@ -61,7 +61,7 @@ def preprocess_imu_data(input_path: Union[str, Path], output_path: Union[str, Pa
 
     # Store data
     for sensor, units in zip(['accelerometer', 'gyroscope'], ['g', config.rotation_units]):
-        df_sensor = df[[config.time_colname] + [x for x in df.columns if sensor in x]]
+        df_sensor = df[[config.time_colname.value] + [x for x in df.columns if sensor in x]]
 
         metadata_samples.channels = [x for x in df.columns if sensor in x]
         metadata_samples.units = list(np.repeat(units, len(metadata_samples.channels)))
@@ -75,8 +75,8 @@ def preprocess_imu_data(input_path: Union[str, Path], output_path: Union[str, Pa
 def transform_time_array(
     time_array: pd.Series,
     scale_factor: float,
-    input_unit_type: str,
-    output_unit_type: str,
+    input_unit_type: TimeUnit,
+    output_unit_type: TimeUnit,
     start_time: float = 0.0,
 ) -> np.ndarray:
     """
@@ -88,9 +88,9 @@ def transform_time_array(
         The time array in milliseconds to transform.
     scale_factor : float
         The scale factor to apply to the time array.
-    input_unit_type : str
+    input_unit_type : TimeUnit
         The time unit type of the input time array. Raw PPP data was in `TimeUnit.difference_ms`.
-    output_unit_type : str
+    output_unit_type : TimeUnit
         The time unit type of the output time array. The processing is often done in `TimeUnit.relative_ms`.
     start_time : float, optional
         The start time of the time array in UNIX milliseconds (default is 0.0)
@@ -129,8 +129,8 @@ def transform_time_array(
 
 def resample_data(
     df: pd.DataFrame,
-    time_column : str,
-    time_unit_type: str,
+    time_column : DataColumns,
+    time_unit_type: TimeUnit,
     unscaled_column_names: List[str],
     resampling_frequency: int,
     scale_factors: List[float] = [],
@@ -177,15 +177,16 @@ def resample_data(
     t_resampled = np.arange(start_time, time_abs_array[-1], 1 / resampling_frequency)
 
     # create dataframe
-    df = pd.DataFrame(t_resampled, columns=[time_column])
+    df = pd.DataFrame(t_resampled, columns=[time_column.value])
 
     # interpolate IMU - maybe a separate method?
     for j, sensor_col in enumerate(unscaled_column_names):
         if not np.all(np.diff(time_abs_array) > 0):
             raise ValueError("time_abs_array is not strictly increasing")
 
         cs = CubicSpline(time_abs_array, scaled_values.T[j])
-        df[sensor_col] = cs(df[time_column])
+        #TODO: isn't sensor_col of type DataColumns?
+        df[sensor_col] = cs(df[time_column.value])
 
     return df
 

src/paradigma/ppg_preprocessing.py

@@ -50,16 +50,16 @@ def preprocess_ppg_data(tsdf_meta_ppg: tsdf.TSDFMetadata, tsdf_meta_imu: tsdf.TS
 
     # Transform the time arrays to absolute milliseconds
     start_time_ppg = parse_iso8601_to_datetime(metadata_time_ppg.start_iso8601).timestamp()
-    df_imu[DataColumns.TIME] = paradigma.imu_preprocessing.transform_time_array(
-        time_array=df_imu[DataColumns.TIME],
+    df_imu[DataColumns.TIME.value] = paradigma.imu_preprocessing.transform_time_array(
+        time_array=df_imu[DataColumns.TIME.value],
         scale_factor=1000, 
         input_unit_type = paradigma.constants.TimeUnit.difference_ms,
         output_unit_type = paradigma.constants.TimeUnit.absolute_ms,
         start_time = start_time_ppg)
 
     start_time_imu = parse_iso8601_to_datetime(metadata_time_imu.start_iso8601).timestamp()
-    df_ppg[DataColumns.TIME] = paradigma.imu_preprocessing.transform_time_array(
-        time_array=df_ppg[DataColumns.TIME],
+    df_ppg[DataColumns.TIME.value] = paradigma.imu_preprocessing.transform_time_array(
+        time_array=df_ppg[DataColumns.TIME.value],
         scale_factor=1000, 
         input_unit_type = paradigma.constants.TimeUnit.difference_ms,
         output_unit_type = paradigma.constants.TimeUnit.absolute_ms,
@@ -122,16 +122,16 @@ def preprocess_ppg_data(tsdf_meta_ppg: tsdf.TSDFMetadata, tsdf_meta_imu: tsdf.TS
             df_ppg_proc = df_ppg_proc.drop(columns=[col])
             df_ppg_proc = df_ppg_proc.rename(columns={f'filt_{col}': col})
 
-    df_imu_proc[DataColumns.TIME] = paradigma.imu_preprocessing.transform_time_array(
-        time_array=df_imu_proc[DataColumns.TIME],
+    df_imu_proc[DataColumns.TIME.value] = paradigma.imu_preprocessing.transform_time_array(
+        time_array=df_imu_proc[DataColumns.TIME.value],
         scale_factor=1,
         input_unit_type=paradigma.constants.TimeUnit.absolute_ms,
         output_unit_type=paradigma.constants.TimeUnit.relative_ms,
         start_time=start_time_ppg,
     )
 
-    df_ppg_proc[DataColumns.TIME] = paradigma.imu_preprocessing.transform_time_array(
-        time_array=df_ppg_proc[DataColumns.TIME],
+    df_ppg_proc[DataColumns.TIME.value] = paradigma.imu_preprocessing.transform_time_array(
+        time_array=df_ppg_proc[DataColumns.TIME.value],
         scale_factor=1,
         input_unit_type=paradigma.constants.TimeUnit.absolute_ms,
         output_unit_type=paradigma.constants.TimeUnit.relative_ms,

src/paradigma/preprocessing_config.py

@@ -37,14 +37,14 @@ def __init__(self) -> None:
         self.rotation_units = 'deg/s'
 
         self.d_channels_accelerometer = {
-            DataColumns.ACCELEROMETER_X: self.acceleration_units,
-            DataColumns.ACCELEROMETER_Y: self.acceleration_units,
-            DataColumns.ACCELEROMETER_Z: self.acceleration_units,
+            DataColumns.ACCELEROMETER_X.value: self.acceleration_units,
+            DataColumns.ACCELEROMETER_Y.value: self.acceleration_units,
+            DataColumns.ACCELEROMETER_Z.value: self.acceleration_units,
         }
         self.d_channels_gyroscope = {
-            DataColumns.GYROSCOPE_X: self.rotation_units,
-            DataColumns.GYROSCOPE_Y: self.rotation_units,
-            DataColumns.GYROSCOPE_Z: self.rotation_units,
+            DataColumns.GYROSCOPE_X.value: self.rotation_units,
+            DataColumns.GYROSCOPE_Y.value: self.rotation_units,
+            DataColumns.GYROSCOPE_Z.value: self.rotation_units,
         }
         self.d_channels_imu = {**self.d_channels_accelerometer, **self.d_channels_gyroscope}
 
@@ -58,7 +58,7 @@ def __init__(self) -> None:
         self.time_filename = 'PPG_time.bin'
 
         self.d_channels_ppg = {
-            DataColumns.PPG: 'none'
+            DataColumns.PPG.value: 'none'
         }
 
         self.sampling_frequency = 30