]ci skip] past irregular

batchalign/pipelines/morphosyntax/ud.py

@@ -248,6 +248,13 @@ def handler__VERB(word, lang=None):
     polarity = feats.get("Polarity", "")
     polite = feats.get("Polite", "")
 
+    irr = False
+    if lang == "en" and tense == "Past":
+        from batchalign.pipelines.morphosyntax.en.irr import is_irregular
+        irr = is_irregular(word.lemma, word.text)
+    irr = "irr" if irr else "" 
+
+
     res = handler(word, lang)
     if "sconj" in res:
         return res
@@ -256,7 +263,7 @@ def handler__VERB(word, lang=None):
     else:
         return res+flag+stringify_feats(aspect, mood,
                                         tense, polarity, polite,
-                                        number[:1]+person)
+                                        number[:1]+person, irr)
 
 def handler__actual_PUNCT(word, lang=None):
     # actual punctuation handler

batchalign/version

@@ -1,3 +1,3 @@
-0.7.6-alpha.18
+0.7.6-alpha.19
 October 16, 2024
-fixing asr for file names
+past irregular

scratchpad.py

@@ -91,7 +91,7 @@
 
 # # from batchalign.models import BertUtteranceModel
 # from batchalign.pipelines import BatchalignPipeline
-# forms, delim = chat_parse_utterance("who is that ?", None, None, None, None)
+# forms, delim = chat_parse_utterance("who farted that ?", None, None, None, None)
 # utterance = Utterance(content=forms, delim=delim)
 # ut = Document(content=[utterance], langs=["eng"])
 # pipe = BatchalignPipeline.new("morphosyntax", "eng")