ChatBot.py

from flask import Flask, render_template, request
from transformers import GPT2LMHeadModel, GPT2Tokenizer
from sklearn.feature_extraction.text import CountVectorizer
from sklearn.metrics.pairwise import cosine_similarity
import json

tokenizer = GPT2Tokenizer.from_pretrained('gpt2')
model = GPT2LMHeadModel.from_pretrained('gpt2')

json_file_path = 'interview-workspace\src\data\short_let_room.json'

app = Flask(__name__)


def generate_gpt2_response(question):
    input_ids = tokenizer.encode(question, return_tensors='pt')
    generated_output = model.generate(input_ids, max_length=len(input_ids[0]) + 100,
                                      num_beams=5,
                                      no_repeat_ngram_size=2,
                                      top_k=10,
                                      top_p=1,
                                      temperature=0.9,
                                      pad_token_id=model.config.eos_token_id)
    generated_response = tokenizer.decode(generated_output[0], skip_special_tokens=True)
    return generated_response

def find_question_and_answer(json_file, question, input_ids):
    with open(json_file, "r") as json_file:
        data = json.load(json_file)

    quest = question.lower()

    all_questions = data.get("questions", [])
    all_response = data.get("responses", [])
    for dataset_question in all_questions:
        if 'question' in dataset_question and dataset_question['question'].lower() == quest:
            for dataset_response in all_response:
                if dataset_question['response_id'] == dataset_response['id']:
                    response = {
                        "Chat Bot ": dataset_response["response_message"],
                        "Apartment ": dataset_response["response_message1"],
                        "Address ": dataset_response["shortlet_Address"],
                        "Price ": dataset_response["shortlet_Price"],
                        "URL ": dataset_response.get("shortlet_url")
                    }
                    return response
                
    input_ids = tokenizer.encode(question, return_tensors='pt')
    generated_output = model.generate(input_ids, max_length=len(input_ids[0]) + 100,
                                    num_beams=5,
                                    no_repeat_ngram_size=2,
                                    top_k=10,
                                    top_p=1,
                                    temperature=0.9,
                                    pad_token_id=model.config.eos_token_id)
    generated_response = tokenizer.decode(generated_output[0], skip_special_tokens=True)
    return generated_response

@app.route("/")
def index():
    return render_template('index.html')

@app.route('/ask', methods=['POST'])
def ask():
    user_input = request.form['user_input']
    if not user_input:
        response = ""
    else:
        input_ids = tokenizer.encode(user_input, return_tensors='pt')
        response = find_question_and_answer(json_file_path, user_input, input_ids)

    if isinstance(response, dict):
        response_type = "mapping"
    else:
        response_type = "string"

    return render_template('index.html', user_input=user_input, response=response, response_type=response_type)


if __name__ == '__main__':
    app.run(debug=True)