Merge pull request #174 from vocalpy/rename-on/offset-ind-to-on/offse…

…t-sample

Rename on/offset ind to on/offset sample

src/crowsetta/formats/seq/birdsongrec.py

@@ -185,8 +185,8 @@ def to_seq(self,
         """
         seqs = []
         for birdsongrec_seq in self.sequences:
-            onset_inds = np.array([syl.position for syl in birdsongrec_seq.syls])
-            offset_inds = np.array([syl.position + syl.length for syl in birdsongrec_seq.syls])
+            onset_samples = np.array([syl.position for syl in birdsongrec_seq.syls])
+            offset_samples = np.array([syl.position + syl.length for syl in birdsongrec_seq.syls])
             labels = np.array(
                 # NOTE we convert syl.label to string so dtype is consistent across formats
                 # and to adhere to schema for `'generic-seq'`
@@ -211,17 +211,17 @@ def to_seq(self,
                 samplerate_this_wav = samplerate
 
             if samplerate_this_wav:
-                onsets_s = onset_inds / samplerate_this_wav
-                offsets_s = offset_inds / samplerate_this_wav
+                onsets_s = onset_samples / samplerate_this_wav
+                offsets_s = offset_samples / samplerate_this_wav
                 if round_times:
                     onsets_s = np.round(onsets_s, decimals=decimals)
                     offsets_s = np.round(offsets_s , decimals=decimals)
             else:
                 onsets_s = None
                 offsets_s = None
 
-            seq = crowsetta.Sequence.from_keyword(onset_inds=onset_inds,
-                                                  offset_inds=offset_inds,
+            seq = crowsetta.Sequence.from_keyword(onset_samples=onset_samples,
+                                                  offset_samples=offset_samples,
                                                   onsets_s=onsets_s,
                                                   offsets_s=offsets_s,
                                                   labels=labels

src/crowsetta/formats/seq/generic.py

@@ -31,7 +31,7 @@
 ONSET_OFFSET_COLS_ERR = """For onset times and offset times, 
 all values must be specified in at least one unit: 
 seconds (float), or sample number (integer). All rows must be non-null for either 
-'onset_s' and 'offset_s' or 'onset_ind' and 'offset_ind'.
+'onset_s' and 'offset_s' or 'onset_sample' and 'offset_sample'.
 Both units can also be specified. Conversion between units is not validated.
 """
 
@@ -44,8 +44,8 @@ class GenericSeqSchema(pandera.SchemaModel):
     label: Series[pd.StringDtype] = pandera.Field(coerce=True)
     onset_s: Optional[Series[float]] = pandera.Field()
     offset_s: Optional[Series[float]] = pandera.Field()
-    onset_ind: Optional[Series[int]] = pandera.Field()
-    offset_ind: Optional[Series[int]] = pandera.Field()
+    onset_sample: Optional[Series[int]] = pandera.Field()
+    offset_sample: Optional[Series[int]] = pandera.Field()
 
     notated_path: Series[str] = pandera.Field()
     annot_path: Series[str] = pandera.Field()
@@ -62,23 +62,23 @@ def both_onset_s_and_offset_s_if_either(cls, df: pd.DataFrame) -> bool:
             return True
 
     @pandera.dataframe_check(error=ONSET_OFFSET_COLS_ERR)
-    def both_onset_ind_and_offset_ind_if_either(cls, df: pd.DataFrame) -> bool:
-        """check that, if one of {'onset_ind', 'offset_ind'} column is present,
+    def both_onset_sample_and_offset_sample_if_either(cls, df: pd.DataFrame) -> bool:
+        """check that, if one of {'onset_sample', 'offset_sample'} column is present,
         then both are present"""
-        if any([col in df for col in ('onset_ind', 'offset_ind')]):
-            return all([col in df for col in ('onset_ind', 'offset_ind')])
+        if any([col in df for col in ('onset_sample', 'offset_sample')]):
+            return all([col in df for col in ('onset_sample', 'offset_sample')])
         else:
             return True
 
     @pandera.dataframe_check(error=ONSET_OFFSET_COLS_ERR)
     def onset_offset_s_and_ind_are_not_both_missing(cls, df: pd.DataFrame) -> bool:
         """check that at least one of the on/offset column pairs is present:
-        either {'onset_s', 'offset_s'} or {'onset_ind', 'offset_ind'}"""
+        either {'onset_s', 'offset_s'} or {'onset_sample', 'offset_sample'}"""
         if 'onset_s' not in df and 'offset_s' not in df:
-            return 'onset_ind' in df and 'offset_ind' in df
-        elif 'onset_ind' not in df and 'offset_ind' not in df:
+            return 'onset_sample' in df and 'offset_sample' in df
+        elif 'onset_sample' not in df and 'offset_sample' not in df:
             return 'onset_s' in df and 'offset_s' in df
-        elif all([col in df for col in ('onset_s', 'offset_s', 'onset_ind', 'offset_ind')]):
+        elif all([col in df for col in ('onset_s', 'offset_s', 'onset_sample', 'offset_sample')]):
             #  i.e., else return True, but extra verbose for clarity
             return True
 
@@ -225,18 +225,18 @@ def csv2annot(csv_path: PathLike) -> List[crowsetta.Annotation]:
         else:
             onsets_s = None
             offsets_s = None
-        if 'onset_ind' and 'offset_ind' in df_annot:
-            onsets_inds = df_annot.onset_ind.values
-            offsets_inds = df_annot.offset_ind.values
+        if 'onset_sample' and 'offset_sample' in df_annot:
+            onsets_inds = df_annot.onset_sample.values
+            offsets_inds = df_annot.offset_sample.values
         else:
             onsets_inds = None
             offsets_inds = None
         seq = crowsetta.Sequence.from_keyword(
             labels=labels,
             onsets_s=onsets_s,
             offsets_s=offsets_s,
-            onset_inds=onsets_inds,
-            offset_inds=offsets_inds,
+            onset_samples=onsets_inds,
+            offset_samples=offsets_inds,
         )
         annot = crowsetta.Annotation(
             annot_path=annot_path,

src/crowsetta/formats/seq/timit.py

@@ -152,8 +152,8 @@ def to_seq(self,
         and then sending them to a csv file,
         the result should be the same on Windows and Linux.
         """
-        onset_inds = self.begin_samples
-        offset_inds = self.end_samples
+        onset_samples = self.begin_samples
+        offset_samples = self.end_samples
         labels = self.text
 
         if samplerate is None:
@@ -169,16 +169,16 @@ def to_seq(self,
                 )
                 samplerate = None
 
-        onsets_s = onset_inds / samplerate
-        offsets_s = offset_inds / samplerate
+        onsets_s = onset_samples / samplerate
+        offsets_s = offset_samples / samplerate
 
         if round_times:
             onsets_s = np.around(onsets_s, decimals=decimals)
             offsets_s = np.around(offsets_s, decimals=decimals)
 
         phn_seq = crowsetta.Sequence.from_keyword(labels=labels,
-                                                  onset_inds=onset_inds,
-                                                  offset_inds=offset_inds,
+                                                  onset_samples=onset_samples,
+                                                  offset_samples=offset_samples,
                                                   onsets_s=onsets_s,
                                                   offsets_s=offsets_s)
         return phn_seq

src/crowsetta/formats/seq/yarden.py

@@ -167,8 +167,8 @@ def to_seq(self,
             )
 
             samp_freq = annotation['fs'].tolist()
-            seq_dict['onset_inds'] = np.round(seq_dict['onsets_s'] * samp_freq).astype(int)
-            seq_dict['offset_inds'] = np.round(seq_dict['offsets_s'] * samp_freq).astype(int)
+            seq_dict['onset_samples'] = np.round(seq_dict['onsets_s'] * samp_freq).astype(int)
+            seq_dict['offset_samples'] = np.round(seq_dict['offsets_s'] * samp_freq).astype(int)
 
             if round_times:
                 seq_dict['onsets_s'] = np.around(seq_dict['onsets_s'], decimals=decimals)

src/crowsetta/segment.py

@@ -26,13 +26,13 @@ class Segment(object):
     used to annotate animal communication.
     Typically, a single unit such as a syllable in human speech
     or a "syllable" in birdsong."""
-    _FIELDS = ('label', 'onset_s', 'offset_s', 'onset_ind', 'offset_ind')
+    _FIELDS = ('label', 'onset_s', 'offset_s', 'onset_sample', 'offset_sample')
 
     label = attr.ib(converter=str)
     onset_s = attr.ib(converter=attr.converters.optional(float_or_None))
     offset_s = attr.ib(converter=attr.converters.optional(float_or_None))
-    onset_ind = attr.ib(converter=attr.converters.optional(int_or_None))
-    offset_ind = attr.ib(converter=attr.converters.optional(int_or_None))
+    onset_sample = attr.ib(converter=attr.converters.optional(int_or_None))
+    offset_sample = attr.ib(converter=attr.converters.optional(int_or_None))
     asdict = attr.asdict
 
     @classmethod
@@ -66,20 +66,20 @@ def from_row(cls, row, header=None):
 
     @classmethod
     def from_keyword(cls, label, onset_s=None, offset_s=None,
-                     onset_ind=None, offset_ind=None):
-        if ((onset_ind is None and offset_ind is None) and
+                     onset_sample=None, offset_sample=None):
+        if ((onset_sample is None and offset_sample is None) and
                 (onset_s is None and offset_s is None)):
-            raise ValueError('must provide either onset_ind and offset_ind, or '
+            raise ValueError('must provide either onset_sample and offset_sample, or '
                              'onsets_s and offsets_s')
 
-        if onset_ind and offset_ind is None:
-            raise ValueError(f'onset_ind specified as {onset_ind} but offset_ind is None')
-        if onset_ind is None and offset_ind:
-            raise ValueError(f'offset_ind specified as {offset_ind} but onset_ind is None')
+        if onset_sample and offset_sample is None:
+            raise ValueError(f'onset_sample specified as {onset_sample} but offset_sample is None')
+        if onset_sample is None and offset_sample:
+            raise ValueError(f'offset_sample specified as {offset_sample} but onset_sample is None')
         if onset_s and offset_s is None:
             raise ValueError(f'onset_s specified as {onset_s} but offset_s is None')
         if onset_s is None and offset_s:
-            raise ValueError(f'offset_s specified as {offset_ind} but onset_s is None')
+            raise ValueError(f'offset_s specified as {offset_sample} but onset_s is None')
 
         return cls(label=label, onset_s=onset_s, offset_s=offset_s,
-                   onset_ind=onset_ind, offset_ind=offset_ind)
+                   onset_sample=onset_sample, offset_sample=offset_sample)