Ground truth now contains column with species id instead of full species description

tools/set_ground_truths_of_reads.py

@@ -26,8 +26,9 @@ def parse_one_xml(file_path):
     """
     species = get_species_from_xml(file_path)
     file_name = file_path.name.replace('_0.xml', '.fast5')
+    species_id = species.split()[0] if species else None
     # print(file_name, species)
-    return file_name, species
+    return file_name, species, species_id
 
 
 def main():
@@ -48,7 +49,8 @@ def main():
     with multiprocessing.Pool(args.num_workers) as p:
         ground_truths = p.map(parse_one_xml, all_files)
     df = pd.DataFrame.from_records(ground_truths, columns=['file name',
-                                                           'species'])
+                                                           'species',
+                                                           'species id'])
     df.to_csv(args.out_path, index=False)