milvas connection

crudmilvus.py

@@ -0,0 +1,12 @@
+import os
+from pymilvus import connections, FieldSchema, CollectionSchema, DataType, Collection
+from sentence_transformers import SentenceTransformer
+
+milvus_token = os.getenv("MILVUS_TOKEN")
+
+COLLECTION_NAME = "final_ragbench_document_embeddings"
+MILVUS_CLOUD_URI = "https://in03-7b4da1b7b588a88.serverless.gcp-us-west1.cloud.zilliz.com"
+connections.connect("default", uri=MILVUS_CLOUD_URI, token=milvus_token)
+
+# Verify connection
+print(connections.get_connection_addr("default"))

loaddataset.py

@@ -86,5 +86,6 @@ def ExtractData(ragbench: Dict[str, dict], split: str = "train") -> pd.DataFrame
 def ExtractRagBenchData():
     ragbench = load_rag_bench_dataset(DATASET_CONFIGS)
     rag_extracted_data = ExtractData(ragbench, split="train")
+    rag_extracted_data["dataset_name"].fillna("covidqa", inplace=True)
 
     return rag_extracted_data