Merge pull request #237 from JohnSnowLabs/release/523/lp-bugfix

Release/523/lp bugfix
JohnSnowLabs · Jan 20, 2024 · 97cd134 · 97cd134
2 parents b392597 + c7572f3
commit 97cd134
Show file tree

Hide file tree

Showing 3 changed files with 5 additions and 3 deletions.
diff --git a/nlu/pipe/col_substitution/col_name_substitution_utils.py b/nlu/pipe/col_substitution/col_name_substitution_utils.py
@@ -113,8 +113,9 @@ def get_final_output_cols_of_component(c, df, anno_2_ex) -> List[str]:
         """Get's a list of all columns that have been derived in the pythonify procedure from the component_to_resolve
         os_components in dataframe df for anno_2_ex configs """
         og_output_col = c.spark_output_column_names[0]
+
         # may be missing because finisher cleaning
-        if og_output_col not in df.columns: return []
+        if og_output_col not in anno_2_ex: return []
         configs = anno_2_ex[og_output_col]
 
         if c.name == NLP_NODE_IDS.FINISHER:

diff --git a/nlu/pipe/extractors/extractor_methods/base_extractor_methods.py b/nlu/pipe/extractors/extractor_methods/base_extractor_methods.py
@@ -21,7 +21,8 @@
 def extract_light_pipe_rows(df):
     """Extract Annotations from Light Pipeline into same represenation as other extractors in thos module"""
     ff = lambda row: list(map(f, row)) if isinstance(row, List) else row
-    f = lambda anno: dict(annotatorType=anno.annotator_type,
+    f = lambda anno: dict(
+                # annotatorType=anno.annotator_type,
                           begin=anno.begin, end=anno.end,
                           result=anno.result,
                           metadata=anno.metadata,

diff --git a/nlu/pipe/utils/predict_helper.py b/nlu/pipe/utils/predict_helper.py
@@ -105,7 +105,7 @@ def predict_multi_threaded_light_pipe(pipe, data, output_level, positions, keep_
     data, stranger_features, output_datatype = DataConversionUtils.to_pandas_df(data, pipe.raw_text_column)
 
     # Predict -> Cast to PDF -> Join with original inputs. It does NOT yield EMBEDDINGS.
-    data = data.join(pd.DataFrame(pipe.light_transformer_pipe.fullAnnotate(data.text.values)))
+    data = data.join(pd.DataFrame(pipe.light_transformer_pipe.fullAnnotate(data.text.values.tolist())))
 
     return pipe.pythonify_spark_dataframe(data,
                                           keep_stranger_features=keep_stranger_features,