mend

app.py

@@ -102,17 +102,28 @@ class EntailmentChecker(BaseComponent):
             # if in the first documents there is a strong evidence of entailment/contradiction,
             # there is no need to consider less relevant documents
             
+        if scores > 0:
+            aggregate_entailment_info = {
+                "contradiction": round(agg_con / scores, 2),
+                "neutral": round(agg_neu / scores, 2),
+                "entailment": round(agg_ent / scores, 2),
+            }
 
-        aggregate_entailment_info = {
-            "contradiction": round(agg_con / scores, 2),
-            "neutral": round(agg_neu / scores, 2),
-            "entailment": round(agg_ent / scores, 2),
-        }
+            entailment_checker_result = {
+                "documents": considered_documents,
+                "aggregate_entailment_info": aggregate_entailment_info,
+            }
+        else:
+            aggregate_entailment_info = {
+                "contradiction": 0,
+                "neutral": 0,
+                "entailment": 0,
+            }
 
-        entailment_checker_result = {
-            "documents": considered_documents[: i + 1],
-            "aggregate_entailment_info": aggregate_entailment_info,
-        }
+            entailment_checker_result = {
+                "documents": considered_documents,
+                "aggregate_entailment_info": aggregate_entailment_info,
+            }
 
         return entailment_checker_result, "output_1"
     

config.py

@@ -1,4 +1,5 @@
-RETRIEVER_MODEL = "sentence-transformers/distiluse-base-multilingual-cased-v1"
+# RETRIEVER_MODEL = "sentence-transformers/distiluse-base-multilingual-cased-v1"
+RETRIEVER_MODEL = "unicamp-dl/mt5-base-en-pt-msmarco-v2"
 RETRIEVER_TOP_K = 15
 
-NLI_MODEL = "microsoft/deberta-large-mnli"
+NLI_MODEL = "MoritzLaurer/mDeBERTa-v3-base-xnli-multilingual-nli-2mil7"