Experiment (2): Drop worst points if no. of points larger than factor…

… times targeted sample size

src/tranquilo/process_arguments.py

@@ -89,6 +89,7 @@ def process_arguments(
     residualize=None,
     # Experimental options
     experimental=None,
+    filter_target_sample_size_factor=None,
 ):
     # warning for things that do not work well yet
     if noisy and functype == "scalar":
@@ -249,6 +250,8 @@ def process_arguments(
         "aggregate_model": aggregate_model,
         "estimate_variance": estimate_variance,
         "accept_candidate": accept_candidate,
+        "filter_target_sample_size_factor": filter_target_sample_size_factor,
+        "experimental": experimental,
     }
 
     return out

src/tranquilo/tranquilo.py

@@ -49,6 +49,9 @@ def _internal_tranquilo(
     aggregate_model,
     estimate_variance,
     accept_candidate,
+    # experimental options
+    experimental,
+    filter_target_sample_size_factor,
 ):
     if n_evals_at_start > 1:
         eval_info = {0: next_multiple(n_evals_at_start, base=batch_size)}
@@ -105,11 +108,18 @@ def _internal_tranquilo(
 
         old_xs = history.get_xs(old_indices)
 
+        if experimental:
+            _filter_target_sample_size_factor = filter_target_sample_size_factor
+        else:
+            _filter_target_sample_size_factor = 1
+
         model_xs, model_indices = filter_points(
             xs=old_xs,
             indices=old_indices,
             state=state,
-            target_size=target_sample_size,
+            target_size=int(
+                np.floor(_filter_target_sample_size_factor * target_sample_size)
+            ),
             history=history,
             n_evals_per_point=n_evals_per_point,
         )