Merge pull request #185 from neo4j-labs/DEV

Staging with gemini

Author: karanchellani

.gitignore

@@ -161,4 +161,5 @@ cython_debug/
 .vscode/launch.json
 temp.pdf
 google-cloud-sdk
-google-cloud-cli-469.0.0-linux-x86_64.tar.gz
+google-cloud-cli-469.0.0-linux-x86_64.tar.gz
+/data/llm-experiments-387609-c73d512ca3b1.json

backend/requirements.txt

@@ -8,8 +8,9 @@ asyncio==3.4.3
 attrs==23.2.0
 backoff==2.2.1
 beautifulsoup4==4.12.3
-boto3==1.34.66
-botocore==1.34.66
+boto3
+botocore
+cachetools==5.3.3
 certifi==2024.2.2
 cffi==1.16.0
 chardet==5.2.0
@@ -23,6 +24,7 @@ dataclasses-json==0.6.4
 dataclasses-json-speakeasy==0.5.11
 Deprecated==1.2.14
 distro==1.9.0
+docstring_parser==0.16
 effdet==0.4.1
 emoji==2.10.1
 exceptiongroup==1.2.0
@@ -34,14 +36,29 @@ flatbuffers==23.5.26
 fonttools==4.49.0
 frozenlist==1.4.1
 fsspec==2024.2.0
+google-api-core==2.18.0
+google-auth==2.29.0
+google-cloud-aiplatform==1.45.0
+google-cloud-bigquery==3.19.0
+google-cloud-core==2.4.1
+google-cloud-resource-manager==1.12.3
+google-cloud-storage==2.16.0
+google-crc32c==1.5.0
+google-resumable-media==2.7.0
+googleapis-common-protos==1.63.0
 greenlet==3.0.3
+grpc-google-iam-v1==0.13.0
+grpcio==1.62.1
+google-ai-generativelanguage
+grpcio-status==1.62.1
 h11==0.14.0
 httpcore==1.0.4
 httpx==0.27.0
 huggingface-hub==0.20.3
 humanfriendly==10.0
 idna==3.6
 importlib-resources==6.1.1
+install==1.3.5
 iopath==0.1.10
 Jinja2==3.1.3
 jmespath==1.0.1
@@ -50,11 +67,13 @@ jsonpatch==1.33
 jsonpath-python==1.0.6
 jsonpointer==2.4
 kiwisolver==1.4.5
-langchain==0.1.12
-langchain-community==0.0.28
-langchain-core==0.1.32
-langchain-experimental==0.0.54
-langchain-openai==0.0.8
+langchain
+langchain-google-genai
+langchain-community
+langchain-core
+langchain-experimental
+langchain-google-vertexai
+langchain-openai
 langchain-text-splitters==0.0.1
 langdetect==1.0.9
 langsmith==0.1.31
@@ -85,7 +104,10 @@ pikepdf==8.11.0
 pillow==10.2.0
 pillow_heif==0.15.0
 portalocker==2.8.2
+proto-plus==1.23.0
 protobuf==4.23.4
+pyasn1==0.6.0
+pyasn1_modules==0.4.0
 pycocotools==2.0.7
 pycparser==2.21
 pydantic==2.6.4
@@ -104,10 +126,12 @@ pytz==2024.1
 PyYAML==6.0.1
 rapidfuzz==3.6.1
 regex==2023.12.25
-requests==2.31.0
+requests
+rsa==4.9
 s3transfer==0.10.1
 safetensors==0.3.2
 scipy==1.10.1
+shapely==2.0.3
 six==1.16.0
 sniffio==1.3.1
 soupsieve==2.5
@@ -122,14 +146,16 @@ tokenizers==0.15.2
 tqdm==4.66.2
 transformers==4.37.1
 triton==2.2.0
+types-protobuf
+types-requests
 typing-inspect==0.9.0
 typing_extensions==4.9.0
 tzdata==2024.1
-unstructured==0.12.6
-unstructured-client==0.18.0
-unstructured-inference==0.7.23
-unstructured.pytesseract==0.3.12
-urllib3==1.26.18
+unstructured
+unstructured-client
+unstructured-inference
+unstructured.pytesseract
+urllib3
 uvicorn==0.29.0
 wikipedia==1.4.0
 wrapt==1.16.0

backend/score.py

@@ -1,12 +1,14 @@
 from fastapi import FastAPI, File, UploadFile, Form
-import uvicorn
-from fastapi import FastAPI, Depends
+from fastapi import FastAPI
 from fastapi_health import health
 from fastapi.middleware.cors import CORSMiddleware
 from src.main import *
+from src.QA_integration import *
+from src.entities.user_credential import user_credential
+import uvicorn
 import asyncio
 import base64
-from src.QA_integration import *
+
 
 
 def healthy_condition():
@@ -61,13 +63,13 @@ async def create_source_knowledge_graph_url(
     database=Form(None),
     aws_access_key_id=Form(None),
     aws_secret_access_key=Form(None),
-    max_limit=Form(5),
-    query_source=Form(None),
     wiki_query=Form(None),
-    model=Form(None)
+    model=Form(None),
+    gcs_bucket_name=Form(None),
+    gcs_bucket_folder=Form(None)
 ):
     return create_source_node_graph_url(
-        uri, userName, password, model, source_url, database, wiki_query, aws_access_key_id, aws_secret_access_key
+        uri, userName, password, model, source_url, database, wiki_query, aws_access_key_id, aws_secret_access_key,gcs_bucket_name, gcs_bucket_folder
     )
 
 
@@ -84,6 +86,9 @@ async def extract_knowledge_graph_from_file(
     aws_secret_access_key=Form(None),
     wiki_query=Form(None),
     max_sources=Form(None),
+    gcs_bucket_name=Form(None),
+    gcs_bucket_folder=Form(None),
+    gcs_blob_filename=Form(None)
 ):
     """
     Calls 'extract_graph_from_file' in a new thread to create Neo4jGraph from a
@@ -108,10 +113,7 @@ async def extract_knowledge_graph_from_file(
             password,
             model,
             database,
-            file=file,
-            source_url=None,
-            wiki_query=wiki_query,
-            max_sources=max_sources,
+            file=file
         )
     elif source_url:
         return await asyncio.to_thread(
@@ -128,7 +130,7 @@ async def extract_knowledge_graph_from_file(
             max_sources=max_sources,
         )
     elif wiki_query:
-         return await asyncio.to_thread(
+        return await asyncio.to_thread(
             extract_graph_from_file,
             uri,
             userName,
@@ -137,9 +139,22 @@ async def extract_knowledge_graph_from_file(
             database,
             wiki_query=wiki_query
         )
+    elif gcs_bucket_name:
+        return await asyncio.to_thread(
+            extract_graph_from_file,
+            uri,
+            userName,
+            password,
+            model,
+            database,
+            gcs_bucket_name = gcs_bucket_name,
+            gcs_bucket_folder = gcs_bucket_folder,
+            gcs_blob_filename = gcs_blob_filename
+        )
+             
 
     else:
-        return {"job_status": "Failure", "error": "No file found"}
+        return {"status": "Failed", "error": "No file found"}
 
 
 @app.get("/sources_list")

backend/src/QA_integration.py

@@ -8,7 +8,6 @@
 from langchain_openai import OpenAIEmbeddings
 import logging
 from langchain_community.chat_message_histories import Neo4jChatMessageHistory
-import asyncio
 load_dotenv()
 
 openai_api_key = os.environ.get('OPENAI_API_KEY')
@@ -30,29 +29,29 @@ def vector_embed_results(qa,question):
 
     return vector_res
 
-def cypher_results(graph,question):
-    cypher_res={}
-    try:
-        graph.refresh_schema()
-        cypher_chain = GraphCypherQAChain.from_llm(
-            graph=graph,
-            cypher_llm=ChatOpenAI(temperature=0, model=model_version),
-            qa_llm=ChatOpenAI(temperature=0, model=model_version),
-            validate_cypher=True, # Validate relationship directions
-            verbose=True,
-            top_k=2
-        )
-        try:
-            cypher_res=cypher_chain.invoke({"query": question})
-        except:
-            cypher_res={}
+# def cypher_results(graph,question):
+#     cypher_res={}
+#     try:
+#         graph.refresh_schema()
+#         cypher_chain = GraphCypherQAChain.from_llm(
+#             graph=graph,
+#             cypher_llm=ChatOpenAI(temperature=0, model=model_version),
+#             qa_llm=ChatOpenAI(temperature=0, model=model_version),
+#             validate_cypher=True, # Validate relationship directions
+#             verbose=True,
+#             top_k=2
+#         )
+#         try:
+#             cypher_res=cypher_chain.invoke({"query": question})
+#         except:
+#             cypher_res={}
 
-    except Exception as e:
-      error_message = str(e)
-      logging.exception(f'Exception in CypherQAChain in QA component:{error_message}')
-    #   raise Exception(error_message)
+#     except Exception as e:
+#       error_message = str(e)
+#       logging.exception(f'Exception in CypherQAChain in QA component:{error_message}')
+#     #   raise Exception(error_message)
 
-    return cypher_res
+#     return cypher_res
 
 def save_chat_history(uri,userName,password,session_id,user_message,ai_message):
     try:
@@ -117,17 +116,19 @@ def QA_RAG(uri,userName,password,question,session_id):
             llm=llm, chain_type="stuff", retriever=neo_db.as_retriever(search_kwargs={'k': 3,"score_threshold": 0.5}), return_source_documents=True
         )
 
-        graph = Neo4jGraph(
-            url=uri,
-            username=userName,
-            password=password
-        )
         vector_res=vector_embed_results(qa,question)
         print('Response from Vector embeddings')
         print(vector_res)
-        cypher_res= cypher_results(graph,question)
-        print('Response from CypherQAChain')
-        print(cypher_res)
+
+        # Disable Cypher Chain QA
+        # graph = Neo4jGraph(
+        #     url=uri,
+        #     username=userName,
+        #     password=password
+        # )
+        # cypher_res= cypher_results(graph,question)
+        # print('Response from CypherQAChain')
+        # print(cypher_res)
 
         chat_summary=get_chat_history(llm,uri,userName,password,session_id)
 
@@ -140,10 +141,11 @@ def QA_RAG(uri,userName,password,question,session_id):
         Given the user's query: {question}, provide a meaningful and efficient answer based
         on the insights derived from the following data:
         chat_summary:{chat_summary}
-        Structured information: {cypher_res.get('result','')}.
+        Structured information:  .
         Unstructured information: {vector_res.get('result','')}.
 
-        """
+        """ 
+
         print(final_prompt)
         response = llm.predict(final_prompt)
         ai_message=response

backend/src/QA_optimization.py

@@ -37,7 +37,7 @@ def __init__(self, uri, userName, password, question, session_id):
     async def execute(self):
         tasks = [
             self._vector_embed_results(),
-            self._cypher_results(),
+            # self._cypher_results(), ## Disabled call for cypher_results
             self._get_chat_history()
         ]
         return await asyncio.gather(*tasks)
@@ -114,8 +114,6 @@ async def _cypher_results(self):
         print("Cypher QA duration",datetime.now()-t)
         return cypher_res
 
-
-
     async def _get_chat_history(self):
         try:
             t=datetime.now()
@@ -195,9 +193,12 @@ async def main(uri,userName,password,question,session_id):
     parallel_component = ParallelComponent(uri, userName, password, question, session_id)
     f_results=await parallel_component.execute()
     print(f_results)
+    # f_vector_result=f_results[0]['result']
+    # f_cypher_result=f_results[1].get('result','')
+    # f_chat_summary=f_results[2]['result']
     f_vector_result=f_results[0]['result']
-    f_cypher_result=f_results[1].get('result','')
-    f_chat_summary=f_results[2]['result']
+    f_cypher_result = "" # Passing Empty string for cypher_result 
+    f_chat_summary=f_results[1]['result']
     print(f_vector_result)
     print(f_cypher_result)
     print(f_chat_summary)

backend/src/api_response.py

@@ -0,0 +1,39 @@
+
+
+def create_api_response(status,success_count=None,Failed_count=None, data=None, error=None,message=None,file_source=None,file_name=None):
+    """
+    Create a response to be sent to the API. This is a helper function to create a JSON response that can be sent to the API.
+    
+    Args:
+        status: The status of the API call. Should be one of the constants in this module.
+        data: The data that was returned by the API call.
+        error: The error that was returned by the API call.
+        success_count: Number of files successfully processed.
+        Failed_count: Number of files failed to process.
+    Returns: 
+      A dictionary containing the status data and error if any
+    """
+    response = {"status": status}
+
+    # Set the data of the response
+    if data is not None:
+      response["data"] = data
+
+    # Set the error message to the response.
+    if error is not None:
+      response["error"] = error
+    
+    if success_count is not None:
+      response['success_count']=success_count
+      response['Failed_count']=Failed_count
+    
+    if message is not None:
+      response['message']=message
+
+    if file_source is not None:
+      response['file_source']=file_source
+
+    if file_name is not None:
+      response['file_name']=file_name
+      
+    return response