from pydantic import NoneStr import os import mimetypes import validators import requests import tempfile import gradio as gr from openai import AzureOpenAI import re import json from transformers import pipeline import matplotlib.pyplot as plt import plotly.express as px import pandas as pd class SentimentAnalyzer: def __init__(self): self.client = AzureOpenAI( api_key = os.getenv("AZURE_OPENAI_API_KEY"), api_version = "2024-02-01", azure_endpoint = os.getenv("AZURE_OPENAI_ENDPOINT") ) def emotion_analysis(self,text): prompt = f""" Your task is find the top 3 emotion for this converstion {text}: and it's emotion score for the Mental Healthcare Doctor Chatbot and patient conversation text.\ you are analyze the text and provide the output in the following list format heigher to lower order: ["emotion1","emotion2","emotion3"][score1,score2,score3]''' [with top 3 result having the highest score] The scores should be in the range of 0.0 to 1.0, where 1.0 represents the highest intensity of the emotion. """ response = self.client.Completion.create( model="text-davinci-003", prompt=prompt, temperature=0, max_tokens=60, top_p=1, frequency_penalty=0, presence_penalty=0 ) message = response.choices[0].text.strip().replace("\n","") return message def analyze_sentiment_for_graph(self, text): prompt = f""" Your task is find the setiments for this converstion {text} : and it's sentiment score for the Mental Healthcare Doctor Chatbot and patient conversation text.\ you are analyze the text and provide the output in the following json format heigher to lower order: '''["label1","label2","label3"][score1,score2,score3]''' """ response = openai.Completion.create( model="text-davinci-003", prompt=prompt, temperature=0, max_tokens=60, top_p=1, frequency_penalty=0, presence_penalty=0 ) # Extract the generated text sentiment_scores = response.choices[0].text.strip() start_index = sentiment_scores.find("[") end_index = sentiment_scores.find("]") list1_text = sentiment_scores[start_index + 1: end_index] list2_text = sentiment_scores[end_index + 2:-1] sentiment = list(map(str.strip, list1_text.split(","))) scores = list(map(float, list2_text.split(","))) score_dict={"Sentiment": sentiment, "Score": scores} print(score_dict) return score_dict def emotion_analysis_for_graph(self,text): start_index = text.find("[") end_index = text.find("]") list1_text = text[start_index + 1: end_index] list2_text = text[end_index + 2:-1] emotions = list(map(str.strip, list1_text.split(","))) scores = list(map(float, list2_text.split(","))) score_dict={"Emotion": emotions, "Score": scores} print(score_dict) return score_dict class Summarizer: def __init__(self): openai.api_key=os.getenv("OPENAI_API_KEY") def generate_summary(self, text): model_engine = "GPT3" prompt = f"""summarize the following conversation delimited by triple backticks. write within 30 words.```{text}``` """ completions = openai.Completion.create( engine=model_engine, prompt=prompt, max_tokens=60, n=1, stop=None, temperature=0.5, ) message = completions.choices[0].text.strip() return message history_state = gr.State() summarizer = Summarizer() sentiment = SentimentAnalyzer() class LangChain_Document_QA: def __init__(self): self.client = AzureOpenAI( api_key = os.getenv("AZURE_OPENAI_API_KEY"), api_version = "2024-02-01", azure_endpoint = os.getenv("AZURE_OPENAI_ENDPOINT") ) def _add_text(self,history, text): history = history + [(text, None)] history_state.value = history return history,gr.update(value="", interactive=False) def _agent_text(self,history, text): response = text history[-1][1] = response history_state.value = history return history def _chat_history(self): history = history_state.value formatted_history = " " for entry in history: customer_text, agent_text = entry formatted_history += f"Patient: {customer_text}\n" if agent_text: formatted_history += f"Mental Healthcare Doctor Chatbot: {agent_text}\n" return formatted_history def _display_history(self): formatted_history=self._chat_history() summary=summarizer.generate_summary(formatted_history) return summary def _display_graph(self,sentiment_scores): df = pd.DataFrame(sentiment_scores) fig = px.bar(df, x='Score', y='Sentiment', orientation='h', labels={'Score': 'Score', 'Labels': 'Sentiment'}) fig.update_layout(height=500, width=200) return fig def _display_graph_emotion(self,customer_emotion_score): fig = px.pie(customer_emotion_score, values='Score', names='Emotion', title='Emotion Distribution', hover_data=['Score']) #fig.update_traces(texttemplate='Emotion', textposition='outside') fig.update_layout(height=500, width=200) return fig def _history_of_chat(self): history = history_state.value formatted_history = "" client="" agent="" for entry in history: customer_text, agent_text = entry client+=customer_text formatted_history += f"Patient: {customer_text}\n" if agent_text: agent+=agent_text formatted_history += f"Mental Healthcare Doctor Chatbot: {agent_text}\n" return client,agent def _suggested_answer(self,history, text): try: history_list = self._chat_history() try: file_path = "patient_details.json" with open(file_path) as file: patient_details = json.load(file) except: pass prompt = f"""Analyse the patient json If asked for information take it from {patient_details} \ you first get patient details : if not match patient json information start new chat else match patient \ json information ask previous: As an empathic AI Mental Healthcare Doctor Chatbot, provide effective solutions to patients' mental health concerns. \ first start the conversation ask existing patient or new patient. if new patient get name,age,gender,contact,address from the patient and start. \ if existing customer get name,age,gender,contact,address details and start the chat about existing issues and current issues. \ if patient say thanking tone message to end the conversation with a thanking greeting when the patient expresses gratitude. \ Chat History:['''{history_list}'''] Patient: ['''{text}'''] Perform as Mental Healthcare Doctor Chatbot """ response = self.client.Completion.create( model="GPT3", prompt=prompt, temperature=0, max_tokens=500, top_p=1, frequency_penalty=0, presence_penalty=0.6, ) message = response.choices[0].text.strip() if ":" in message: message = re.sub(r'^.*:', '', message) history[-1][1] = message.strip() history_state.value = history return history except: history[-1][1] = "How can I help you?" history_state.value = history return history def _text_box(self,customer_emotion,customer_sentiment_score): sentiment_str = ', '.join([f'{label}: {score}' for label, score in zip(customer_sentiment_score['Sentiment'], customer_sentiment_score['Score'])]) #emotion_str = ', '.join([f'{emotion}: {score}' for emotion, score in zip(customer_emotion['Emotion'], customer_emotion['Score'])]) return f"Sentiment: {sentiment_str},\nEmotion: {customer_emotion}" def _on_sentiment_btn_click(self): client=self._history_of_chat() customer_emotion=sentiment.emotion_analysis(client) customer_sentiment_score = sentiment.analyze_sentiment_for_graph(client) scores=self._text_box(customer_emotion,customer_sentiment_score) customer_fig=self._display_graph(customer_sentiment_score) customer_fig.update_layout(title="Sentiment Analysis",width=800) customer_emotion_score = sentiment.emotion_analysis_for_graph(customer_emotion) customer_emotion_fig=self._display_graph_emotion(customer_emotion_score) customer_emotion_fig.update_layout(title="Emotion Analysis",width=800) return scores,customer_fig,customer_emotion_fig def clear_func(self): history_state.clear() def gradio_interface(self): with gr.Blocks(css="style.css",theme='freddyaboulton/test-blue') as demo: with gr.Row(): gr.HTML("""

ADOPLE AI


AI Mental Healthcare ChatBot

""") with gr.Row(): with gr.Column(scale=1): with gr.Row(): chatbot = gr.Chatbot([], elem_id="chatbot").style(height=360) with gr.Row(): with gr.Column(scale=0.90): txt = gr.Textbox(show_label=False,placeholder="Patient").style(container=False) with gr.Column(scale=0.10): emptyBtn = gr.Button("๐Ÿงน Clear") with gr.Accordion("Conversational AI Analytics", open = False): with gr.Row(): with gr.Column(scale=0.50): txt4 =gr.Textbox( show_label=False, lines=4, placeholder="Summary").style(container=False) with gr.Column(scale=0.50): txt5 =gr.Textbox( show_label=False, lines=4, placeholder="Sentiment").style(container=False) with gr.Row(): with gr.Column(scale=0.50, min_width=0): end_btn=gr.Button(value="End") with gr.Column(scale=0.50, min_width=0): Sentiment_btn=gr.Button(value="๐Ÿ“Š",callback=self._on_sentiment_btn_click) with gr.Row(): gr.HTML("""

Sentiment and Emotion Score Graph

""") with gr.Row(): with gr.Column(scale=1, min_width=0): plot =gr.Plot(label="Patient", size=(500, 600)) with gr.Row(): with gr.Column(scale=1, min_width=0): plot_3 =gr.Plot(label="Patient_Emotion", size=(500, 600)) txt_msg = txt.submit(self._add_text, [chatbot, txt], [chatbot, txt]).then( self._suggested_answer, [chatbot,txt],chatbot) txt_msg.then(lambda: gr.update(interactive=True), None, [txt]) # txt.submit(self._suggested_answer, [chatbot,txt],chatbot) # button.click(self._agent_text, [chatbot,txt3], chatbot) end_btn.click(self._display_history, [], txt4) emptyBtn.click(self.clear_func,[],[]) emptyBtn.click(lambda: None, None, chatbot, queue=False) Sentiment_btn.click(self._on_sentiment_btn_click,[],[txt5,plot,plot_3]) demo.title = "AI Mental Healthcare ChatBot" demo.launch() document_qa =LangChain_Document_QA() document_qa.gradio_interface()