Shrikrishna's picture
Update app.py
53f2ac0
raw
history blame contribute delete
No virus
3.34 kB
import openai
import json
import os
from dotenv import load_dotenv
#from pyairtable import Table
import requests
import streamlit as st
load_dotenv()
openai.api_key = os.getenv("OPENAI_API_KEY")
rapid_api_key = os.getenv("X-RapidAPI-Key")
#airtable_api_key = os.getenv("AIRTABLE_API_KEY")
#table = Table(airtable_api_key, "appHojHIE4y8gVBgc", "tbldUUKZFngr78ogg")
function_descriptions = [
{
"name": "get_stock_movers",
"description": "Get the stocks that has biggest price/volume moves, e.g. actives, gainers, losers, etc.",
"parameters": {
"type": "object",
"properties": {
},
}
},
{
"name": "get_stock_news",
"description": "Get the latest news for a stock",
"parameters": {
"type": "object",
"properties": {
"performanceId": {
"type": "string",
"description": "id of the stock, which is referred as performanceID in the API"
},
},
"required": ["performanceId"]
}
},
{
"name": "add_stock_news_airtable",
"description": "Add the stock, news summary & price move to Airtable",
"parameters": {
"type": "object",
"properties": {
"stock": {
"type": "string",
"description": "stock ticker"
},
"move": {
"type": "string",
"description": "price move in %"
},
"news_summary": {
"type": "string",
"description": "news summary of the stock"
},
}
}
},
]
def get_stock_news(performanceId):
url = "https://morning-star.p.rapidapi.com/news/list"
querystring = {"performanceId":performanceId}
headers = {
"X-RapidAPI-Key": rapid_api_key,
"X-RapidAPI-Host": "morning-star.p.rapidapi.com"
}
response = requests.get(url, headers=headers, params=querystring)
short_news_list = response.json()[:5]
print("response:", response, " json response:", short_news_list)
return short_news_list
def get_stock_movers():
url = "https://morning-star.p.rapidapi.com/market/v2/get-movers"
headers = {
"X-RapidAPI-Key": rapid_api_key,
"X-RapidAPI-Host": "morning-star.p.rapidapi.com"
}
response = requests.get(url, headers=headers)
return response.json()
def function_call(ai_response):
function_call = ai_response["choices"][0]["message"]["function_call"]
function_name = function_call["name"]
arguments = function_call["arguments"]
if function_name == "get_stock_movers":
return get_stock_movers()
elif function_name == "get_stock_news":
performanceId = eval(arguments).get("performanceId")
return get_stock_news(performanceId)
else:
return
query = "Give me a summary of what happend to the tesla stocks today?"
messages = [{"role":"user","content":query}]
response = openai.ChatCompletion.create(
model="gpt-3.5-turbo-0613",
messages=messages,
functions = function_descriptions,
function_call="auto"
)
st.subheader(response)