savvyyy/Aspect-Intent-Sentiment-Analysis ? reverse-engineered prompt

Build me a small NLP app that can analyze text and tweets in a few different ways from one local server. I want an API that can take a sentence or post and return the overall sentiment, the polarity, whether it is subjective, and a readable label like very positive, positive, neutral, negative, or very negative. I also want an aspect based mode for domains like restaurant, laptop, and Twitter, where it can find the main aspects being talked about and tell me the sentiment for each one. Add an intent endpoint too, so it can guess what the user is trying to do, like complain, buy, or sell. There should also be a graph endpoint that gives date wise tweet polarity so I can plot trends over time. Make it easy to run locally with a simple setup, using the existing notebooks and model files if they are already there, and have the app start on localhost so I can test the endpoints in Postman. If anything is missing, look up the current docs online and fill in the gaps.