Updated the ads index with healthline and webmd urls ads, also picked only health category ads. System prompt updated to remove relation check

.gitignore

@@ -0,0 +1 @@
+**/.DS_Store

data/142_adclick_20May_20Jun_webmd_healthline_Health_dupRemoved0.8_someAdsCampaign.tsv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ce108af93e6d9924fd80e20d19e86918995e98310c2528021a14ab4b7edd5ee3
+size 9099622

rag.py

@@ -12,8 +12,8 @@ os.environ["TOKENIZERS_PARALLELISM"] = "false"
 
 load_dotenv(override=True)
 client = OpenAI()
-DB_FAISS_PATH = "./vectorstore/db_faiss_ads_1month"
-data_file_path = "./data/134_ads_data_1May_1June_perform.tsv"
+DB_FAISS_PATH = "./vectorstore/db_faiss_ads_20May_20Jun_webmd_healthline_Health_dupRemoved0.8"
+data_file_path = "./data/142_adclick_20May_20Jun_webmd_healthline_Health_dupRemoved0.8_someAdsCampaign.tsv"
 embedding_model_hf = "BAAI/bge-m3"
 qa_model_name = "gpt-3.5-turbo"
 default_threshold = 0.95
@@ -86,6 +86,11 @@ Got A Rosemary Bush? Here’re 20 Brilliant & Unusual Ways To Use All That Rosem
 }
 -----------------------------------------------
 
+The ADS_DATA provided to you is as follows:
+"""
+
+old_system_prompt_additional_example = """
+-----------------------------------------------
 <Sample INPUT(PAGE_TITLE)> 
 7 Signs and Symptoms of Magnesium Deficiency
 
@@ -101,8 +106,6 @@ Ad 3: Weak or Paralyzed Muscles? - A Common Symptom of Cataplexy. About 70% of P
 "options": []
 }
 ------------------------------------------------
-
-The ADS_DATA provided to you is as follows:
 """
 
 embeddings_hf = HuggingFaceEmbeddings(model_name=embedding_model_hf)
@@ -253,7 +256,7 @@ db = FAISS.load_local(
     DB_FAISS_PATH, embeddings_hf, allow_dangerous_deserialization=True
 )
 data = pd.read_csv(data_file_path, sep="\t")
-data.dropna(axis=0, how="any", inplace=True)
+# data.dropna(axis=0, how="any", inplace=True)
 data.drop_duplicates(subset=["ad_title", "ad_desc"], inplace=True)
 ad_title_content = list(data["ad_title"].values)
 with gr.Blocks() as demo:

vectorstore/db_faiss_ads_20May_20Jun_webmd_healthline_Health_dupRemoved0.8/index.faiss

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4a945f8a29f3870a2a3970936089dc3d7ee0ccba9a7ffaa485c6bb29b3a05387
+size 61648941

vectorstore/db_faiss_ads_20May_20Jun_webmd_healthline_Health_dupRemoved0.8/index.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:625301b20eaf328fca1d4eab90912e2688c7637cfefa40f32cb195febd53a0ec
+size 6307355