Creating the Generative Tax Bot: using the pre-trained model from huggieface which Bart model to generate the reponses for the user queries
MAIN.py
the main.py runs the main code this will connect the telegram which chatbot will run on the telegram via api
Crete_faiss.py
this will create the embeddings and files what has to be created
embeddings.py
this is the engine of the chatbot this will help in handling the chatbot working nature which includes Bart model, responsible for generating the chat responses
webscrape.py
this will scrap the data from the url sites and outputs the data into txt and i stored the data into the Doc folder
i tested the responses questions.txt which basically has only few information, from that information i tested in telegram
#Running the Bot
first you need to place you telegram token in main.py and for that generate the token from the telegram from BotFather channel
after placing your token, just run the main.py file
Then go to telegram start testing !!!!!!! :)