Bugfix in evaluate

src/robofish/evaluate/app.py

@@ -98,7 +98,7 @@ def evaluate(args=None):
         args = parser.parse_args()
 
     if args.analysis_type == "all" and args.save_path is None:
-        raise Exception("When the analysis type is all, a path must be given.")
+        raise Exception("When the analysis type is all, a --save_path must be given.")
 
     if args.analysis_type in fdict:
         if args.labels is None:

src/robofish/evaluate/evaluate.py

@@ -60,7 +60,6 @@ def evaluate_speed(
 
         # Exclude possible nans
         path_speeds = np.array(path_speeds)
-        path_speeds = path_speeds[~np.isnan(path_speeds)]
 
         left_quantiles.append(np.quantile(path_speeds, 0.001))
         right_quantiles.append(np.quantile(path_speeds, 0.999))
@@ -117,9 +116,7 @@ def evaluate_turn(
             for e_speeds_turns in speeds_turns:
                 path_turns.extend(np.rad2deg(e_speeds_turns[:, 1]))
 
-        # Exclude possible nans
         path_turns = np.array(path_turns)
-        path_turns = path_turns[~np.isnan(path_turns)]
 
         left_quantiles.append(np.quantile(path_turns, 0.001))
         right_quantiles.append(np.quantile(path_turns, 0.999))
@@ -393,8 +390,8 @@ def evaluate_tank_position(
         new_xy_positions = np.concatenate(
             [p[:, ::poses_step, :2].reshape(-1, 2) for p in poses_per_path], axis=0
         )
-        # Exclude possible nans
-        xy_positions.append(new_xy_positions[~np.isnan(new_xy_positions).any(axis=1)])
+
+        xy_positions.append(new_xy_positions)
 
     fig, ax = plt.subplots(1, len(xy_positions), figsize=(8 * len(xy_positions), 8))
     if len(xy_positions) == 1:
@@ -466,11 +463,7 @@ def evaluate_social_vector(
                             )
                         )
 
-        # Concatenate and exclude possible nans
-        path_socialVec = np.concatenate(path_socialVec, axis=0)
-        path_socialVec = path_socialVec[~np.isnan(path_socialVec).any(axis=1)]
-
-        socialVec.append(path_socialVec)
+        socialVec.append(np.concatenate(path_socialVec, axis=0))
 
     grids = []
 
@@ -535,30 +528,37 @@ def evaluate_follow_iid(
                             calculate_follow(poses[i, :, :2], poses[j, :, :2])
                         )
 
-        follow.append(np.array(path_follow))
-        iid.append(np.array(path_iid))
+        follow.append(np.concatenate([np.atleast_1d(f) for f in path_follow]))
+        iid.append(np.concatenate([np.atleast_1d(i) for i in path_iid]))
 
     grids = []
 
-    follow_flat = np.concatenate([f.flatten() for f in follow])
-    iid_flat = np.concatenate([i.flatten() for i in iid])
-
     # Exclude possible nans
-    follow_flat = follow_flat[~np.isnan(follow_flat)]
-    iid_flat = iid_flat[~np.isnan(iid_flat)]
 
     # Find the 0.5%/ 99.5% quantiles as min and max for the axis
     follow_range = np.max(
-        [-1 * np.quantile(follow_flat, 0.005), np.quantile(follow_flat, 0.995)]
+        [
+            -1 * np.quantile(np.concatenate(follow), 0.1),
+            np.quantile(np.concatenate(follow), 0.9),
+        ]
     )
-    max_iid = np.quantile(iid_flat, 0.995)
+    max_iid = np.quantile(np.concatenate(iid, dtype=np.float32), 0.9)
 
     for i in range(len(follow)):
+
+        # Mask data which is outside of the ranges
+        mask = (
+            (follow[i] > -1 * follow_range)
+            & (follow[i] < follow_range)
+            & (iid[i] < max_iid)
+        )
+
         follow_iid_data = pd.DataFrame(
             {
-                "IID [cm]": np.concatenate(iid[i], axis=0),
-                "Follow": np.concatenate(follow[i], axis=0),
-            }
+                "IID [cm]": iid[i][mask],
+                "Follow": follow[i][mask],
+            },
+            dtype=np.float32,
         )
 
         plt.rcParams["lines.markersize"] = 1
@@ -569,9 +569,10 @@ def evaluate_follow_iid(
             kind="hist",
             xlim=(0, max_iid),
             ylim=(-follow_range, follow_range),
-            cbar=True,
-            legend=True,
-            joint_kws={"bins": 50},
+            # cbar=True,
+            # legend=True,
+            joint_kws={"bins": 30},
+            marginal_kws=dict(bins=30),
         )
         # grid.fig.set_figwidth(9)
         # grid.fig.set_figheight(6)
@@ -623,7 +624,7 @@ def evaluate_tracks(
     file_settings=None,
     lw_distances=False,
     seed=42,
-    max_timesteps=4000,
+    max_timesteps=None,
 ):
     """Evaluate the track.
 

src/robofish/io/utils.py

@@ -104,6 +104,10 @@ def get_all_data_from_paths(
                 data = file.select_entity_property(
                     pred, entity_property=properties[request_type]
                 )
+
+                # Exclude timesteps where there is any nan in the row
+                data = data[:, ~np.isnan(data).any(axis=2).any(axis=0)]
+
                 data_from_files.append(data)
 
         all_data.append(data_from_files)