feat(evaluation): Add decimals parameter to score_predictions. Make tools calling score_predictions default to 3 decimals

Author: nmaarnio

eis_toolkit/cli.py

@@ -2375,7 +2375,7 @@ def classifier_test_cli(
         predictions, reference_profile["height"], reference_profile["width"], nodata_mask
     )
 
-    metrics_dict = score_predictions(y, predictions, get_enum_values(test_metrics))
+    metrics_dict = score_predictions(y, predictions, get_enum_values(test_metrics), decimals=3)
     typer.echo("Progress: 80%")
 
     out_profile = reference_profile.copy()
@@ -2421,7 +2421,7 @@ def regressor_test_cli(
         predictions, reference_profile["height"], reference_profile["width"], nodata_mask
     )
 
-    metrics_dict = score_predictions(y, predictions, get_enum_values(test_metrics))
+    metrics_dict = score_predictions(y, predictions, get_enum_values(test_metrics), decimals=3)
     typer.echo("Progress: 80%")
 
     out_profile = reference_profile.copy()
@@ -3340,6 +3340,7 @@ def score_predictions_cli(
     true_labels: INPUT_FILE_OPTION,
     predictions: INPUT_FILE_OPTION,
     metrics: Annotated[List[str], typer.Option()],
+    decimals: Optional[int] = None,
 ):
     """Score predictions."""
     from eis_toolkit.evaluation.scoring import score_predictions
@@ -3350,7 +3351,7 @@ def score_predictions_cli(
     (y_pred, y_true), _, _ = read_data_for_evaluation([predictions, true_labels])
     typer.echo("Progress: 25%")
 
-    outputs = score_predictions(y_true, y_pred, metrics)
+    outputs = score_predictions(y_true, y_pred, metrics, decimals)
     typer.echo("Progress: 100% \n")
 
     typer.echo(f"Results: {str(outputs)}")

eis_toolkit/evaluation/scoring.py

@@ -1,4 +1,5 @@
 from numbers import Number
+from typing import Optional
 
 import numpy as np
 import pandas as pd
@@ -19,7 +20,10 @@
 
 @beartype
 def score_predictions(
-    y_true: Union[np.ndarray, pd.Series], y_pred: Union[np.ndarray, pd.Series], metrics: Union[str, Sequence[str]]
+    y_true: Union[np.ndarray, pd.Series],
+    y_pred: Union[np.ndarray, pd.Series],
+    metrics: Union[str, Sequence[str]],
+    decimals: Optional[int] = None,
 ) -> Union[Number, Dict[str, Number]]:
     """
     Score model predictions with given metrics.
@@ -34,18 +38,20 @@ def score_predictions(
         y_pred: Predicted labels.
         metrics: The metrics to use for scoring the model. Select only metrics applicable
             for the model type.
+        decimals: Number of decimals used in rounding the scores. If None, scores are not rounded.
+            Defaults to None.
 
     Returns:
         Metric scores as a dictionary if multiple metrics, otherwise just the metric value.
     """
     if isinstance(metrics, str):
         score = _score_predictions(y_true, y_pred, metrics)
-        return score
+        return round(score, decimals) if decimals is not None else score
     else:
         out_metrics = {}
         for metric in metrics:
             score = _score_predictions(y_true, y_pred, metric)
-            out_metrics[metric] = score
+            out_metrics[metric] = round(score, decimals) if decimals is not None else score
         return out_metrics
 
 

eis_toolkit/prediction/machine_learning_general.py

@@ -350,7 +350,7 @@ def _train_and_validate_sklearn_model(
 
         out_metrics = {}
         for metric in metrics:
-            score = score_predictions(y_valid, y_pred, metric)
+            score = score_predictions(y_valid, y_pred, metric, decimals=3)
             out_metrics[metric] = score
 
     # Validation approach 3: Cross-validation
@@ -369,7 +369,7 @@ def _train_and_validate_sklearn_model(
             y_pred = model.predict(X[valid_index])
 
             for metric in metrics:
-                score = score_predictions(y[valid_index], y_pred, metric)
+                score = score_predictions(y[valid_index], y_pred, metric, decimals=3)
                 all_scores = out_metrics[metric][f"{metric}_all"]
                 all_scores.append(score)