Browse Source

fix nltk averaged_perceptron_tagger download and fix score limit is none (#7582)

Co-authored-by: crazywoola <100913391+crazywoola@users.noreply.github.com>
Jyong 7 months ago
parent
commit
7ae728a9a3
3 changed files with 3 additions and 1 deletions
  1. 1 0
      api/.idea/vcs.xml
  2. 1 1
      api/Dockerfile
  3. 1 0
      api/core/rag/retrieval/dataset_retrieval.py

+ 1 - 0
api/.idea/vcs.xml

@@ -12,5 +12,6 @@
   </component>
   <component name="VcsDirectoryMappings">
     <mapping directory="" vcs="Git" />
+    <mapping directory="$PROJECT_DIR$/.." vcs="Git" />
   </component>
 </project>

+ 1 - 1
api/Dockerfile

@@ -65,7 +65,7 @@ COPY --from=packages ${VIRTUAL_ENV} ${VIRTUAL_ENV}
 ENV PATH="${VIRTUAL_ENV}/bin:${PATH}"
 
 # Download nltk data
-RUN python -c "import nltk; nltk.download('punkt')"
+RUN python -c "import nltk; nltk.download('punkt'); nltk.download('averaged_perceptron_tagger')"
 
 # Copy source code
 COPY . /app/api/

+ 1 - 0
api/core/rag/retrieval/dataset_retrieval.py

@@ -616,6 +616,7 @@ class DatasetRetrieval:
         for document in all_documents:
             if score_threshold is None or document.metadata['score'] >= score_threshold:
                 filter_documents.append(document)
+
         if not filter_documents:
             return []
         filter_documents = sorted(filter_documents, key=lambda x: x.metadata['score'], reverse=True)