Create app.py

app.py

@@ -0,0 +1,44 @@
+from haystack.document_stores import InMemoryDocumentStore
+import pandas as pd
+import gradio as gr
+
+df=pd.read_parquet('df.parquet')
+candidats=pd.read_parquet('candidats.parquet')
+
+document_store = InMemoryDocumentStore(use_bm25=True)
+docs=df.drop_duplicates(subset=['fileclean']).rename(columns={'fileclean':'content'}).to_dict(orient='records')
+document_store.write_documents(docs)
+from haystack.nodes import BM25Retriever
+retriever = BM25Retriever(document_store=document_store)
+from haystack.pipelines import DocumentSearchPipeline
+pipeline = DocumentSearchPipeline(retriever=retriever)
+
+def semanticsearch(query):
+    result = pipeline.run(
+          query=query,
+          params={
+              "Retriever": {
+                  "top_k": 10
+              }
+          },debug=False
+      )
+    results=[]
+    for document in result['documents']:
+        result=document.to_dict()
+        for c in ['content_type','embedding','id']:
+          result.pop(c)
+        results.append(result)
+    results=pd.DataFrame(results)
+    return results
+
+demo = gr.Interface(
+    semanticsearch,
+    [
+        gr.Dropdown([candidats.sort_values(by='text').text.tolist()]),
+    ],
+    [gr.Dataframe()]
+    
+)
+
+if __name__ == "__main__":
+    demo.launch()