bernard-ng/drc-ners-nlp
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

feat: fix annotated datatype

Browse Source
This commit is contained in:
bernard-ng
2025-07-24 17:17:52 +02:00
parent 9f410ca674
commit cbe3b0ecf2
1 changed files with 4 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files
processing/annotate.py
+4 -1
View File
@@ -67,7 +67,10 @@ def build_updates(llm_model: str, df: pd.DataFrame, entries: pd.DataFrame) -> pd
if idx % BATCH_SIZE == 0 or idx == len(entries): if idx % BATCH_SIZE == 0 or idx == len(entries):
df.update(pd.DataFrame.from_dict(dict(updates), orient="index")) update_df = pd.DataFrame.from_dict(dict(updates), orient="index")
update_df = update_df['annotated'].astype('Int8').fillna(0)
df.update(update_df)
save_checkpoint(df) save_checkpoint(df)
updates.clear() # avoid re-applying same updates updates.clear() # avoid re-applying same updates