Propagating changes in the prefsampling (#80)

* Propagate changes in prefsampling

* Black formatting

abcvoting/generate.py

@@ -354,7 +354,12 @@ def random_resampling_profile(num_voters, num_cand, p, phi):
     """
     return prefsampling_wrapper(
         app_samplers.resampling,
-        {"num_voters": num_voters, "num_candidates": num_cand, "phi": phi, "p": p},
+        {
+            "num_voters": num_voters,
+            "num_candidates": num_cand,
+            "phi": phi,
+            "rel_size_central_vote": p,
+        },
     )
 
 
@@ -411,8 +416,8 @@ def random_disjoint_resampling_profile(num_voters, num_cand, p, phi=None, num_gr
             "num_voters": num_voters,
             "num_candidates": num_cand,
             "phi": phi,
-            "p": p,
-            "g": num_groups,
+            "rel_size_central_vote": p,
+            "num_central_votes": num_groups,
         },
     )
 
@@ -462,13 +467,13 @@ def random_noise_model_profile(num_voters, num_cand, p, phi, distance="hamming")
     """
 
     if distance == "hamming":
-        noise_type = app_samplers.NoiseType.HAMMING
+        noise_type = app_samplers.SetDistance.HAMMING
     elif distance == "jaccard":
-        noise_type = app_samplers.NoiseType.JACCARD
+        noise_type = app_samplers.SetDistance.JACCARD
     elif distance == "zelinka":
-        noise_type = app_samplers.NoiseType.ZELINKA
+        noise_type = app_samplers.SetDistance.ZELINKA
     elif distance == "bunke-shearer":
-        noise_type = app_samplers.NoiseType.BUNKE_SHEARER
+        noise_type = app_samplers.SetDistance.BUNKE_SHEARER
     else:
         raise ValueError(f"unknown name of distance: {distance}")
     return prefsampling_wrapper(
@@ -477,8 +482,8 @@ def random_noise_model_profile(num_voters, num_cand, p, phi, distance="hamming")
             "num_voters": num_voters,
             "num_candidates": num_cand,
             "phi": phi,
-            "p": p,
-            "noise_type": noise_type,
+            "rel_size_central_vote": p,
+            "distance": noise_type,
         },
     )
 
@@ -542,24 +547,27 @@ def prefsampling_euclidean_wrapper(
     -------
         abcvoting.preferences.Profile | tuple[abcvoting.preferences.Profile,np.ndarray,np.ndarray]
     """
-    voter_prob_distribution_func = None
-    if voter_prob_distribution is not None:
-        voter_prob_distribution_func = voter_prob_distribution.prefsampling_function()
-    candidate_prob_distribution_func = None
-    if candidate_prob_distribution is not None:
-        candidate_prob_distribution_func = candidate_prob_distribution.prefsampling_function()
+    num_dimensions = -1
+    voters_positions = None
+    if voter_points is None:
+        if voter_prob_distribution is not None:
+            voters_positions = voter_prob_distribution.prefsampling_function()
+            num_dimensions = int(voter_prob_distribution.name[0])
+    else:
+        voters_positions = voter_points
+        num_dimensions = len(voter_points[0])
+    candidates_positions = None
+    if candidate_points is None:
+        if candidate_prob_distribution is not None:
+            candidates_positions = candidate_prob_distribution.prefsampling_function()
+    else:
+        candidates_positions = candidate_points
     voters_pos, candidates_pos = prefsampling_sample_election_positions(
-        num_voters,
-        num_cand,
-        voter_prob_distribution_func,
-        dict(),
-        candidate_prob_distribution_func,
-        dict(),
-        voter_points,
-        candidate_points,
+        num_voters, num_cand, num_dimensions, voters_positions, candidates_positions
     )
     sampler_params["num_voters"] = num_voters
     sampler_params["num_candidates"] = num_cand
+    sampler_params["num_dimensions"] = num_dimensions
     sampler_params["voters_positions"] = voters_pos
     sampler_params["candidates_positions"] = candidates_pos
     profile = prefsampling_wrapper(sampler, sampler_params)
@@ -896,35 +904,35 @@ def __init__(self, name, center_point=0.5, sigma=0.15, width=1.0):
 
     def prefsampling_function(self):
         if self.name == "1d_interval":
-            return lambda num_points: point_samplers.cube(
+            return lambda num_points, num_dimensions: point_samplers.cube(
                 num_points, 1, center_point=self.center_point, widths=self.width
             )
         elif self.name == "1d_gaussian":
-            return lambda num_points: point_samplers.gaussian(
+            return lambda num_points, num_dimensions: point_samplers.gaussian(
                 num_points, 1, center_point=self.center_point, sigmas=self.sigma
             )
         elif self.name == "1d_gaussian_interval":
-            return lambda num_points: point_samplers.gaussian(
+            return lambda num_points, num_dimensions: point_samplers.gaussian(
                 num_points,
                 1,
                 center_point=self.center_point,
                 sigmas=self.sigma,
                 widths=[self.width],
             )
         elif self.name == "2d_disc":
-            return lambda num_points: point_samplers.ball_uniform(
+            return lambda num_points, num_dimensions: point_samplers.ball_uniform(
                 num_points, 2, center_point=self.center_point, widths=self.width
             )
         elif self.name == "2d_square":
-            return lambda num_points: point_samplers.cube(
+            return lambda num_points, num_dimensions: point_samplers.cube(
                 num_points, 2, center_point=self.center_point, widths=self.width
             )
         elif self.name == "2d_gaussian":
-            return lambda num_points: point_samplers.gaussian(
+            return lambda num_points, num_dimensions: point_samplers.gaussian(
                 num_points, 2, center_point=self.center_point, sigmas=self.sigma
             )
         elif self.name == "2d_gaussian_disc":
-            return lambda num_points: point_samplers.ball_resampling(
+            return lambda num_points, num_dimensions: point_samplers.ball_resampling(
                 num_points,
                 2,
                 lambda: point_samplers.gaussian(
@@ -935,7 +943,7 @@ def prefsampling_function(self):
                 width=self.width,
             )
         elif self.name == "3d_cube":
-            return lambda num_points: point_samplers.cube(
+            return lambda num_points, num_dimensions: point_samplers.cube(
                 num_points, 3, center_point=self.center_point, widths=self.width
             )
         else:

examples/preflib-files/new_example.cat

@@ -1,7 +1,7 @@
-# FILE NAME: /home/martin/Documents/code/pycharm-projects/abcvoting/examples/preflib-files/new_example.cat
+# FILE NAME: C:\Users\reysi\PycharmProjects\abcvoting\examples\preflib-files/new_example.cat
 # TITLE: 
 # DESCRIPTION: 
-# DATA TYPE: 
+# DATA TYPE: cat
 # MODIFICATION TYPE: 
 # RELATES TO: 
 # RELATED FILES: 

setup.py

@@ -92,7 +92,7 @@ def read_version():
         "numpy>=1.21",
         "gurobipy>=11.0",
         "preflibtools>=2.0.12",
-        "prefsampling>=0.1.15",
+        "prefsampling>=0.1.18",
     ],
     extras_require={
         "dev": [

tests/data/test5.cat

@@ -1,7 +1,7 @@
-# FILE NAME: /home/martin/Documents/code/pycharm-projects/abcvoting/tests/data/test5.cat
+# FILE NAME: C:\Users\reysi\PycharmProjects\abcvoting\tests/data/test5.cat
 # TITLE: 
 # DESCRIPTION: 
-# DATA TYPE: 
+# DATA TYPE: cat
 # MODIFICATION TYPE: 
 # RELATES TO: 
 # RELATED FILES: