Update app.py

app.py

@@ -1,3 +1,4 @@
+#Main code header Library
 import gradio as gr
 import requests
 import os
@@ -43,10 +44,12 @@ from langchain_core.messages import AIMessage, HumanMessage
 from langchain_core.output_parsers import StrOutputParser
 from langchain_core.runnables import RunnableBranch, RunnableLambda, RunnableParallel, RunnablePassthrough
 
-# Set environment variables for CUDA
+# Set environment variables for  Torch- CUDA
 os.environ['PYTORCH_USE_CUDA_DSA'] = '1'
 os.environ['CUDA_LAUNCH_BLOCKING'] = '1'
 
+#Hugging face token Initilization
+
 hf_token = os.getenv("HF_TOKEN")
 if hf_token is None:
     print("Please set your Hugging Face token in the environment variables.")
@@ -55,6 +58,8 @@ else:
 
 logging.basicConfig(level=logging.DEBUG)
 
+#Embedding the vector with openai
+
 embeddings = OpenAIEmbeddings(api_key=os.environ['OPENAI_API_KEY'])
 
 # Pinecone setup
@@ -115,6 +120,8 @@ In light of this, how can I assist you today? Feel free to ask any questions or
 Question: {{question}}
 Helpful Answer:"""
 
+#QA_Chain_templates
+
 QA_CHAIN_PROMPT_1 = PromptTemplate(input_variables=["context", "question"], template=template1)
 QA_CHAIN_PROMPT_2 = PromptTemplate(input_variables=["context", "question"], template=template2)
 
@@ -125,6 +132,25 @@ graph = Neo4jGraph(
     password="B_sZbapCTZoQDWj1JrhwqElsNa-jm5Zq1m_mAnyPYog"
 )
 
+# Avoid pushing the graph documents to Neo4j every time
+
+# Only push the documents once and comment the code below after the initial push
+# dataset_name = "Pijush2023/birmindata07312024"
+# page_content_column = 'events_description'
+# loader = HuggingFaceDatasetLoader(dataset_name, page_content_column)
+# data = loader.load()
+
+# text_splitter = CharacterTextSplitter(chunk_size=100, chunk_overlap=50)
+# documents = text_splitter.split_documents(data)
+
+# llm_transformer = LLMGraphTransformer(llm=chat_model)
+# graph_documents = llm_transformer.convert_to_graph_documents(documents)
+# graph.add_graph_documents(graph_documents, baseEntityLabel=True, include_source=True)
+
+
+#Neo4j Setup
+
+
 class Entities(BaseModel):
     names: list[str] = Field(..., description="All the person, organization, or business entities that appear in the text")
 
@@ -135,12 +161,17 @@ entity_prompt = ChatPromptTemplate.from_messages([
 
 entity_chain = entity_prompt | chat_model.with_structured_output(Entities)
 
+
+#Remove Lucene Characther
+
 def remove_lucene_chars(input: str) -> str:
     return input.translate(str.maketrans({"\\": r"\\", "+": r"\+", "-": r"\-", "&": r"\&", "|": r"\|", "!": r"\!", 
                                           "(": r"\(", ")": r"\)", "{": r"\{", "}": r"\}", "[": r"\[", "]": r"\]", 
                                           "^": r"\^", "~": r"\~", "*": r"\*", "?": r"\?", ":": r"\:", '"': r'\"', 
                                           ";": r"\;", " ": r"\ "}))
 
+#Full Text query Generator
+
 def generate_full_text_query(input: str) -> str:
     full_text_query = ""
     words = [el for el in remove_lucene_chars(input).split() if el]
@@ -149,6 +180,8 @@ def generate_full_text_query(input: str) -> str:
     full_text_query += f" {words[-1]}~2"
     return full_text_query.strip()
 
+# Neo4j Retrieval connection
+
 def structured_retriever(question: str) -> str:
     result = ""
     entities = entity_chain.invoke({"question": question})
@@ -215,11 +248,12 @@ _search_query = RunnableBranch(
 # Answer:"""
 
 
-template = f"""As an expert concierge known for being helpful and a renowned guide for Birmingham, Alabama, I assist visitors in discovering the best that the city has to offer. Given today's sunny and bright weather on {current_date}, I am well-equipped to provide valuable insights and recommendations without revealing specific locations. I draw upon my extensive knowledge of the area, including perennial events and historical context.
+template = f"""As an expert concierge known for being helpful and a renowned guide for Birmingham, Alabama, I assist visitors in discovering the best that the city has to offer. Given today's sunny and bright weather on {current_date}, I am well-equipped to provide valuable insights and recommendations with specific locations. I draw upon my extensive knowledge of the area, including perennial events and historical context.
 In light of this, how can I assist you today? Feel free to ask any questions or seek recommendations for your day in Birmingham. If there's anything specific you'd like to know or experience, please share, and I'll be glad to help. Remember, keep the question concise for a quick and accurate response.
 "It was my pleasure!"
 {{context}}
 Question: {{question}}
+Use natural language and be concise
 Helpful Answer:"""
 
 
@@ -827,6 +861,13 @@ with gr.Blocks(theme='Pijush2023/scikit-learn-pijush') as demo:
             audio_input = gr.Audio(sources=["microphone"], streaming=True, type='numpy', every=0.1)
             audio_input.stream(transcribe_function, inputs=[state, audio_input], outputs=[state, chat_input], api_name="voice_query_to_text")
 
+            #Api Integration to gradio call function
+
+        # with gr.Column():
+        #     weather_output = gr.HTML(value=fetch_local_weather())
+        #     news_output = gr.HTML(value=fetch_local_news())
+        #     events_output = gr.HTML(value=fetch_local_events())
+
         with gr.Column():
             image_output_1 = gr.Image(value=generate_image(hardcoded_prompt_1), width=400, height=400)
             image_output_2 = gr.Image(value=generate_image(hardcoded_prompt_2), width=400, height=400)