Hi TN: fraction class improvements with lexical mappings for special cases

Jenkinsfile

@@ -28,7 +28,7 @@ pipeline {
     MR_TN_CACHE='/home/jenkins/TestData/text_norm/ci/grammars/03-12-24-1'
     JA_TN_CACHE='/home/jenkins/TestData/text_norm/ci/grammars/10-17-24-1'
     KO_TN_CACHE='/home/jenkins/TestData/text_norm/ci/grammars/04-23-26-0'
-    HI_TN_CACHE='/home/jenkins/TestData/text_norm/ci/grammars/04-23-26-0'
+    HI_TN_CACHE='/home/jenkins/TestData/text_norm/ci/grammars/04-28-26-0'
     DEFAULT_TN_CACHE='/home/jenkins/TestData/text_norm/ci/grammars/06-08-23-0'
   }
   stages {

nemo_text_processing/text_normalization/hi/taggers/fraction.py

@@ -32,20 +32,6 @@
 
 
 class FractionFst(GraphFst):
-    """
-    Finite state transducer for classifying fraction
-    "२३ ४/६" ->
-    fraction { integer: "तेईस" numerator: "चार" denominator: "छः"}
-    ४/६" ->
-    fraction { numerator: "चार" denominator: "छः"}
-
-
-    Args:
-        cardinal: cardinal GraphFst
-        deterministic: if True will provide a single transduction option,
-            for False multiple transduction are generated (used for audio-based normalization)
-    """
-
     def __init__(self, cardinal, deterministic: bool = True):
         super().__init__(name="fraction", kind="classify", deterministic=deterministic)
 
@@ -54,15 +40,20 @@ def __init__(self, cardinal, deterministic: bool = True):
         self.optional_graph_negative = pynini.closure(
             pynutil.insert("negative: ") + pynini.cross("-", "\"true\"") + pynutil.insert(NEMO_SPACE), 0, 1
         )
+
         self.integer = pynutil.insert("integer_part: \"") + cardinal_graph + pynutil.insert("\"")
+
         self.numerator = (
             pynutil.insert("numerator: \"")
             + cardinal_graph
             + pynini.cross(pynini.union("/", NEMO_SPACE + "/" + NEMO_SPACE), "\"")
             + pynutil.insert(NEMO_SPACE)
         )
+
         self.denominator = pynutil.insert("denominator: \"") + cardinal_graph + pynutil.insert("\"")
 
+        # ---------------- EXISTING SPECIAL FORMS ----------------
+
         dedh_dhai_graph = pynini.string_map(
             [("१" + NEMO_SPACE + HI_ONE_HALF, HI_DEDH), ("२" + NEMO_SPACE + HI_ONE_HALF, HI_DHAI)]
         )
@@ -77,6 +68,27 @@ def __init__(self, cardinal, deterministic: bool = True):
         paune_numbers = paune + pynini.cross(NEMO_SPACE + HI_THREE_QUARTERS, "")
         paune_graph = pynutil.insert(HI_PAUNE) + pynutil.insert(NEMO_SPACE) + paune_numbers
 
+        # ---------------- COMMON FRACTION FORMS ----------------
+
+        common_fraction_map = pynini.string_map(
+            [
+                ("१/२", "आधा"),
+                ("१/३", "तिहाई"),
+                ("२/३", "दो तिहाई"),
+                ("१/४", "चौथाई"),
+                ("३/४", "तीन चौथाई"),
+            ]
+        )
+
+        graph_common_fraction = (
+            pynutil.insert("morphosyntactic_features: \"")
+            + common_fraction_map
+            + pynutil.insert("\"")
+            + pynutil.insert(NEMO_SPACE)
+        )
+
+        # ---------------- WRAPPING GRAPHS ----------------
+
         graph_dedh_dhai = (
             pynutil.insert("morphosyntactic_features: \"")
             + dedh_dhai_graph
@@ -105,19 +117,24 @@ def __init__(self, cardinal, deterministic: bool = True):
             + pynutil.insert(NEMO_SPACE)
         )
 
+        # ---------------- DEFAULT FRACTION ----------------
+
         final_graph = (
             self.optional_graph_negative
             + pynini.closure(self.integer + pynini.accep(NEMO_SPACE), 0, 1)
             + self.numerator
             + self.denominator
         )
 
+        # ---------------- PRIORITY HANDLING ----------------
+
         weighted_graph = (
             final_graph
+            | pynutil.add_weight(graph_common_fraction, -0.3)  # ensures override of "बटा"
             | pynutil.add_weight(graph_dedh_dhai, -0.2)
+            | pynutil.add_weight(graph_paune, -0.2)
             | pynutil.add_weight(graph_savva, -0.1)
             | pynutil.add_weight(graph_sadhe, -0.1)
-            | pynutil.add_weight(graph_paune, -0.2)
         )
 
         self.graph = weighted_graph