Load tables from files

Author: Marcin Kardas

extract_tables.py

@@ -276,6 +276,16 @@ def save_tables(data, outdir):
         json.dump(metadata, f)
 
 
+def load_tables(path):
+    path = Path(path)
+    with open(path / "metadata.json", "r") as f:
+        metadata = json.load(f)
+
+    return [Table.from_file(
+        path,
+        table_metadata) for table_metadata in metadata]
+
+
 def set_ids_by_labels(soup):
     captions = soup.select(".ltx_caption")
     for caption in captions:

sota_extractor2/models/linking/linker.py

@@ -17,14 +17,10 @@ def __call__(self, paper, tables, topk=1):
                                      dataset_extractor=self.dataset_extractor,
                                      topk=topk)
 
-        if topk == 1:
-            proposals = proposals.set_index('cell_ext_id')
-            best = proposals
-        else:
-            best = self.get_best_proposals(proposals)
+        proposals = proposals.set_index('cell_ext_id')
 
-        pipeline_logger(f"{Linker.step}::linked", paper=paper, tables=tables, proposals=best)
+        pipeline_logger(f"{Linker.step}::linked", paper=paper, tables=tables, proposals=proposals)
         return proposals
 
     def get_best_proposals(self, proposals):
-        return proposals.groupby('cell_ext_id').head(1).set_index('cell_ext_id')
+        return proposals.groupby('cell_ext_id').head(1)