Update main.py

main.py

@@ -1,63 +1,57 @@
-from fastapi import FastAPI
-import pickle
-import uvicorn
-import pandas as pd
+from fastapi import FastAPI, Request
+from fastapi.responses import JSONResponse
+from pydantic import BaseModel
+
+from langchain import PromptTemplate, LLMChain
+from langchain.memory import StreamlitChatMessageHistory
+import numpy as np
+from langchain.chains import LLMChain
+from langchain.prompts import PromptTemplate
+from langchain.memory import ConversationBufferMemory
+from langchain.memory.chat_message_histories import StreamlitChatMessageHistory
+from langchain import HuggingFaceHub
+import os
+from dotenv import load_dotenv
+load_dotenv()
+from pathlib import Path
+from huggingface_hub import InferenceClient
+from langchain import HuggingFaceHub
+import requests
+import uuid
+import sys
+
+
+HUGGINGFACEHUB_API_TOKEN = os.getenv('HUGGINGFACEHUB_API_TOKEN')
+repo_id = os.environ.get('repo_id')
+port = os.getenv('port')
+
+llm = HuggingFaceHub(repo_id=repo_id,
+                     model_kwargs={"min_length":1024,
+                                   "max_new_tokens":5632, "do_sample":True,
+                                   "temperature":0.1,
+                                   "top_k":50,
+                                   "top_p":0.95, "eos_token_id":49155}) 
+
+prompt_template = """
+<<SYS>>You are a helpful, respectful and honest assistant. Always answer as helpfully as possible, while being safe.  Your answers should not include any harmful, unethical, racist, sexist, toxic, dangerous, or illegal content. Please ensure that your responses are socially unbiased and positive in nature.
+If a question does not make any sense, or is not factually coherent, explain why instead of answering something not correct. If you don't know the answer to a question, please don't share false information.
+In each conversation, question is placed after [INST] while your answer should be placed after [/INST].<</SYS>>
+[INST] {user_question} [/INST]
+assistant:
+"""
+
+llm_chain = LLMChain(llm=llm, prompt=PromptTemplate.from_template(prompt_template))
 
 app = FastAPI()
 
-# @app.get("/")
-# def read_root():
-#     return {"Hello": "World!"}
-
-
-# Function to load pickle file
-def load_pickle(filename):
-    with open(filename, 'rb') as file:
-        data = pickle.load(file)
-        return data
-    
-# Load pickle file
-ml_components = load_pickle('ml_sepsis.pkl') 
-
-# Components in the pickle file
-ml_model = ml_components['model']
-pipeline_processing = ml_components['pipeline']    
-
-#Endpoints 
-#Root endpoints
-@app.get("/")
-def root():
-    return {"API": "An API for Sepsis Prediction."}
-    
-@app.get('/Predict_Sepsis')
-async def predict(Plasma_glucose: int, Blood_Work_Result_1: int,
-                  Blood_Pressure: int, Blood_Work_Result_2: int,
-                    Blood_Work_Result_3: int, Body_mass_index: float, 
-                    Blood_Work_Result_4: float,Age: int, Insurance:float):
-    
-    data = pd.DataFrame({'Plasma glucose': [Plasma_glucose], 'Blood Work Result-1':	[Blood_Work_Result_1],
-                         'Blood Pressure': [Blood_Pressure], 'Blood Work Result-2': [Blood_Work_Result_2],
-                        'Blood Work Result-3': [Blood_Work_Result_3], 'Body mass index': [Body_mass_index],
-                        'Blood Work Result-4':	[Blood_Work_Result_4], 'Age': [Age], 'Insurance':[Insurance]})
-    
-    data_prepared = pipeline_processing.transform(data)
-
-    model_output = ml_model.predict(data_prepared).tolist()
-
-    prediction = make_prediction(model_output)
-
-    return prediction
-    
-    
-
-
-def make_prediction(data_prepared):
-
-    output_pred = data_prepared
-
-    if output_pred == 0:
-        output_pred = "Sepsis status is Negative"
-    else:
-        output_pred = "Sepsis status is Positive"
-        
-    return output_pred
+class ChatRequest(BaseModel):
+    user_question: str
+
+@app.post('/api/chat')
+async def chat(request: Request, chat_request: ChatRequest):
+    user_query = chat_request.user_question
+    initial_response = llm_chain.run(user_query)
+    return JSONResponse({'response': initial_response})
+
+if __name__ == '__main__':
+    uvicorn.run(app, host='0.0.0.0')