test

main.py

@@ -15,6 +15,8 @@ import textwrap
 from flask_cors import CORS
 import socket;
 
+import gradio as gr
+
 app = Flask(__name__)
 cors = CORS(app)
 
@@ -66,40 +68,45 @@ def default():
     return "Hello World!"
 
 
-if __name__ == '__main__':
-    ip=get_local_ip()
-    os.environ["OPENAI_API_KEY"] = "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"
-    # Embed and store the texts
-    # if(torch.cuda.is_available() == False):
-    #     print("No GPU available")
-    #     exit(1)
+# if __name__ == '__main__':
+#     ip=get_local_ip()
+#     os.environ["OPENAI_API_KEY"] = "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"
+#     # Embed and store the texts
+#     # if(torch.cuda.is_available() == False):
+#     #     print("No GPU available")
+#     #     exit(1)
     
-    torch.cuda.empty_cache()
-    torch.max_split_size_mb = 100
-    instructor_embeddings = HuggingFaceInstructEmbeddings(model_name="hkunlp/instructor-xl", 
-                                                      model_kwargs={"device": "cpu"})
-    # Supplying a persist_directory will store the embeddings on disk
-    persist_directory = 'db'
-    vectordb2 = Chroma(persist_directory=persist_directory, 
-                  embedding_function=instructor_embeddings,
-                   )
-    retriever = vectordb2.as_retriever(search_kwargs={"k": 3})
-    vectordb2.persist()
+#     torch.cuda.empty_cache()
+#     torch.max_split_size_mb = 100
+#     instructor_embeddings = HuggingFaceInstructEmbeddings(model_name="hkunlp/instructor-xl", 
+#                                                       model_kwargs={"device": "cpu"})
+#     # Supplying a persist_directory will store the embeddings on disk
+#     persist_directory = 'db'
+#     vectordb2 = Chroma(persist_directory=persist_directory, 
+#                   embedding_function=instructor_embeddings,
+#                    )
+#     retriever = vectordb2.as_retriever(search_kwargs={"k": 3})
+#     vectordb2.persist()
+
+#     # Set up the turbo LLM
+#     turbo_llm = ChatOpenAI(
+#         temperature=0,
+#         model_name='gpt-3.5-turbo'
+#     )
+#     qa_chain = RetrievalQA.from_chain_type(llm=turbo_llm, 
+#                                   chain_type="stuff", 
+#                                   retriever=retriever, 
+#                                   return_source_documents=True)
+#     qa_chain.combine_documents_chain.llm_chain.prompt.messages[0].prompt.template= """
+#     Use only the following pieces of context and think step by step to answer. Answer the users question only if they are related to the context given.
+#     If you don't know the answer, just say that you don't know, don't try to make up an answer. Make your answer very detailed and long. 
+#     Use bullet points to explain when required. 
+#     Use only text found in the context as your knowledge source for the answer. 
+#     ----------------
+#     {context}"""
+#     app.run(host=ip, port=5000)
 
-    # Set up the turbo LLM
-    turbo_llm = ChatOpenAI(
-        temperature=0,
-        model_name='gpt-3.5-turbo'
-    )
-    qa_chain = RetrievalQA.from_chain_type(llm=turbo_llm, 
-                                  chain_type="stuff", 
-                                  retriever=retriever, 
-                                  return_source_documents=True)
-    qa_chain.combine_documents_chain.llm_chain.prompt.messages[0].prompt.template= """
-    Use only the following pieces of context and think step by step to answer. Answer the users question only if they are related to the context given.
-    If you don't know the answer, just say that you don't know, don't try to make up an answer. Make your answer very detailed and long. 
-    Use bullet points to explain when required. 
-    Use only text found in the context as your knowledge source for the answer. 
-    ----------------
-    {context}"""
-    app.run(host=ip, port=5000)
+def greet(name):
+    return "Hello " + name + "!!"
+iface = gr.Interface(fn=greet, inputs="text", outputs="text")
+iface.launch()