iiegn Claude Sonnet 4.5 commited on
Commit
611853b
·
verified ·
1 Parent(s): 40b471f

Critical fix: MWT detection was incorrectly including empty nodes

Browse files

Issue: Empty nodes (8, '.', 1) were being added to BOTH mwts and empty_nodes:
- As MWT: "8-1" (wrong - invalid ID)
- As empty_node: "8.1" (correct)

Root cause: MWT detection checked isinstance(token['id'], tuple) without
checking the middle element. Both MWTs and empty nodes are tuples:
- MWT: (1, '-', 2) for "1-2"
- Empty node: (8, '.', 1) for "8.1"

Fix: Add token['id'][1] == '-' check to MWT detection to exclude empty nodes

This caused 75 validation errors in sentences with empty nodes.

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>

tools/04_generate_parquet.py CHANGED
@@ -90,9 +90,11 @@ def extract_examples_from_conllu(filepath: str) -> List[Dict[str, Any]]:
90
  sent_idx = str(idx)
91
 
92
  # Extract Multi-Word Tokens (MWTs) - tokens with tuple IDs like (1, '-', 2)
 
93
  mwts = []
94
  for token in sent:
95
- if isinstance(token["id"], tuple): # MWT line (e.g., "1-2")
 
96
  mwts.append({
97
  "id": f"{token['id'][0]}-{token['id'][2]}",
98
  "form": token["form"],
 
90
  sent_idx = str(idx)
91
 
92
  # Extract Multi-Word Tokens (MWTs) - tokens with tuple IDs like (1, '-', 2)
93
+ # Note: Exclude empty nodes which have '.' as middle element: (22, '.', 1)
94
  mwts = []
95
  for token in sent:
96
+ if isinstance(token["id"], tuple) and len(token["id"]) == 3 and token["id"][1] == '-':
97
+ # MWT line (e.g., (1, '-', 2) for "1-2")
98
  mwts.append({
99
  "id": f"{token['id'][0]}-{token['id'][2]}",
100
  "form": token["form"],
tools/templates/universal_dependencies.tmpl CHANGED
@@ -194,9 +194,11 @@ class UniversalDependencies(datasets.GeneratorBasedBuilder):
194
  idx = id
195
 
196
  # Extract Multi-Word Tokens (MWTs) - tokens with tuple IDs like (1, '-', 2)
 
197
  mwts = []
198
  for token in sent:
199
- if isinstance(token["id"], tuple): # MWT line (e.g., "1-2")
 
200
  mwts.append({
201
  "id": f"{token['id'][0]}-{token['id'][2]}",
202
  "form": token["form"],
 
194
  idx = id
195
 
196
  # Extract Multi-Word Tokens (MWTs) - tokens with tuple IDs like (1, '-', 2)
197
+ # Note: Exclude empty nodes which have '.' as middle element: (22, '.', 1)
198
  mwts = []
199
  for token in sent:
200
+ if isinstance(token["id"], tuple) and len(token["id"]) == 3 and token["id"][1] == '-':
201
+ # MWT line (e.g., (1, '-', 2) for "1-2")
202
  mwts.append({
203
  "id": f"{token['id'][0]}-{token['id'][2]}",
204
  "form": token["form"],