basic classes

pyproject.toml

@@ -10,7 +10,7 @@ packages = [{include = "new_fave", from="src"}]
 [tool.poetry.dependencies]
 python = ">=3.11,<3.12"
 aligned-textgrid = "^0.6.3"
-fasttrackpy = "^0.3.0"
+fasttrackpy = "^0.4.0"
 numpy = "^1.26.4"
 
 

src/new_fave/measurements/vowel_measurement.py

@@ -0,0 +1,89 @@
+from fasttrackpy import CandidateTracks, OneTrack
+from aligned_textgrid import AlignedTextGrid
+from collections import defaultdict
+import numpy as np
+
+import warnings
+
+class VowelClassCollection(defaultdict):
+    def __init__(self, track_list:list[CandidateTracks]):
+        super().__init__(lambda : VowelClass())
+        self.tracks_dict = defaultdict(lambda: [])
+        self._make_tracks_dict(track_list)
+        self._dictify()
+
+
+
+    def __setitem__(self, __key, __value) -> None:
+        super().__setitem__(__key, __value)
+
+    def _make_tracks_dict(self, track_list):
+        for v in track_list:
+            self.tracks_dict[v.label].append(v)
+
+    def _dictify(self):
+        for v in self.tracks_dict:
+            self[v] = VowelClass(v, self.tracks_dict[v])
+
+class VowelClass():
+    def __init__(
+            self,
+            label: str,
+            tracks: list     
+        ):
+        self.label = label
+        self.tracks = tracks
+        self._winners = [x.winner for x in self.tracks]
+
+    @property
+    def winners(self):
+        return self._winners
+
+    @winners.setter
+    def winners(self, idces):
+        self._winners = [
+            t.candidates[idx] 
+            for t, idx in zip(self.tracks, idces)
+        ]
+
+    @property
+    def winner_params(self):
+        first_param = np.vstack(
+            [
+                x.parameters[:,0] 
+                for x in self.winners
+            ]
+        ).T
+
+        return first_param
+
+    @property
+    def params_means(self):
+        return self.winner_params.mean(axis = 1)
+
+    @property
+    def params_covs(self):
+        with warnings.catch_warnings():
+            warnings.simplefilter("ignore")
+            param_cov = np.cov(self.winner_params)
+        return param_cov
+
+class VowelMeasurement():
+    def __init__(
+            self, 
+            track: CandidateTracks
+        ):
+        self.track = track
+
+    @property
+    def vowel_class(self):
+        if self._vclass:
+            return self._vclass
+
+    @vowel_class.setter
+    def vowel_class(self, vclass: VowelClass):
+        self._vclass = vclass
+
+
+
+