feat: decades, serial weight fix, 451/470 (96%)

- Date: add decades pattern (nineteen eighties => 1980s)
- Telephone: increase serial weight to reduce false matches
- Telephone: add double digit support in IP
- Update test cases (451 cases)

Author: pengzhendong

itn/english/rules/date.py

@@ -118,7 +118,14 @@ def build_tagger(self):
         # Year only => "twenty twelve", "two thousand three"
         graph_y = add_weight(year, 0.01) + po
 
-        final_graph = graph_mdy | graph_md | graph_my | graph_dmy | graph_dm | graph_y
+        # Decades: "nineteen eighties" => 1980s
+        decade_suffix = closure(self.ALPHA, 1) + (cross("ies", "y") | delete("s"))
+        decade_word = pynini.compose(decade_suffix, ties | cross("ten", "10"))
+        graph_decade = (
+            insert('year: "') + (teen | two_digit) + ds + decade_word + insert('0s"') + po
+        )
+
+        final_graph = graph_mdy | graph_md | graph_my | graph_dmy | graph_dm | graph_y | graph_decade
         self.tagger = self.add_tokens(final_graph)
 
     def build_verbalizer(self):

itn/english/rules/telephone.py

@@ -87,7 +87,7 @@ def build_tagger(self):
         serial_char = single | add_weight(two_digit, 0.002) | self.ALPHA
         serial = serial_char + closure(ds + serial_char, 2)
         serial = serial @ (closure(self.ALPHA | self.DIGIT) + self.DIGIT + closure(self.ALPHA | self.DIGIT))
-        graph |= insert('number_part: "') + add_weight(serial, 0.001) + insert('"')
+        graph |= insert('number_part: "') + add_weight(serial, 2.0) + insert('"')
 
         self.tagger = self.add_tokens(graph)
 

itn/english/test/data/en_date.txt

@@ -1,4 +1,5 @@
 july twenty fifth two thousand twelve => july 25 2012
+nineteen eighties => 1980s
 two thousand and twenty => 2020
 two thousand and nine => 2009
 the twenty fifth of july twenty twelve => 25 july 2012