app.py

import gradio as gr
from transformers import pipeline
import os
import pandas as pd
import numpy as np
import joblib
import spacy
from langchain_core.pydantic_v1 import BaseModel, Field
from langchain.prompts import HumanMessagePromptTemplate, ChatPromptTemplate
from langchain.output_parsers import PydanticOutputParser
from langchain_openai import ChatOpenAI

# Set up models for each app
chat = ChatOpenAI()
classifier = pipeline("sentiment-analysis", model="cardiffnlp/twitter-xlm-roberta-base-sentiment")
asr = pipeline("automatic-speech-recognition", "facebook/wav2vec2-base-960h")
summarizer = pipeline("summarization", model="knkarthick/MEETING_SUMMARY")
fin_model = pipeline("sentiment-analysis", model='yiyanghkust/finbert-tone', tokenizer='yiyanghkust/finbert-tone')
fls_model = pipeline("text-classification", model="demo-org/finbert_fls", tokenizer="demo-org/finbert_fls")

# --- Translator App ---
class TextTranslator(BaseModel):
    output: str = Field(description="Python string containing the output text translated in the desired language")

output_parser = PydanticOutputParser(pydantic_object=TextTranslator)
format_instructions = output_parser.get_format_instructions()

def text_translator(input_text : str, language : str) -> str:
    human_template = """Enter the text that you want to translate:
    {input_text}, and enter the language that you want it to translate to {language}. {format_instructions}"""
    human_message_prompt = HumanMessagePromptTemplate.from_template(human_template)
    chat_prompt = ChatPromptTemplate.from_messages([human_message_prompt])
    prompt = chat_prompt.format_prompt(input_text = input_text, language = language, format_instructions = format_instructions)
    messages = prompt.to_messages()
    response = chat(messages = messages)
    output = output_parser.parse(response.content)
    output_text = output.output
    return output_text

# --- Sentiment Analysis App ---
def sentiment_analysis(message, history):
    result = classifier(message)
    return f"Sentimiento : {result[0]['label']} (Probabilidad: {result[0]['score']:.2f})"

# --- Financial Analyst App ---
nlp = spacy.load('en_core_web_sm')
nlp.add_pipe('sentencizer')

def split_in_sentences(text):
    doc = nlp(text)
    return [str(sent).strip() for sent in doc.sents]

def make_spans(text, results):
    results_list = [results[i]['label'] for i in range(len(results))]
    return list(zip(split_in_sentences(text), results_list))

def summarize_text(text):
    resp = summarizer(text)
    return resp[0]['summary_text']

def text_to_sentiment(text):
    sentiment = fin_model(text)[0]["label"]
    return sentiment

def fin_ext(text):
    results = fin_model(split_in_sentences(text))
    return make_spans(text, results)

def fls(text):
    results = fls_model(split_in_sentences(text))
    return make_spans(text, results)

# --- Customer Churn App ---
script_dir = os.path.dirname(os.path.abspath(__file__))
pipeline_path = os.path.join(script_dir, 'toolkit', 'pipeline.joblib')
model_path = os.path.join(script_dir, 'toolkit', 'Random Forest Classifier.joblib')

pipeline = joblib.load(pipeline_path)
model = joblib.load(model_path)

def calculate_total_charges(tenure, monthly_charges):
    return tenure * monthly_charges

def predict(SeniorCitizen, Partner, Dependents, tenure, InternetService, OnlineSecurity, OnlineBackup, DeviceProtection,
            TechSupport, StreamingTV, StreamingMovies, Contract, PaperlessBilling, PaymentMethod, MonthlyCharges):
    TotalCharges = calculate_total_charges(tenure, MonthlyCharges)
    input_df = pd.DataFrame({
        'SeniorCitizen': [SeniorCitizen],
        'Partner': [Partner],
        'Dependents': [Dependents],
        'tenure': [tenure],
        'InternetService': [InternetService],
        'OnlineSecurity': [OnlineSecurity],
        'OnlineBackup': [OnlineBackup],
        'DeviceProtection': [DeviceProtection],
        'TechSupport': [TechSupport],
        'StreamingTV': [StreamingTV],
        'StreamingMovies': [StreamingMovies],
        'Contract': [Contract],
        'PaperlessBilling': [PaperlessBilling],
        'PaymentMethod': [PaymentMethod],
        'MonthlyCharges': [MonthlyCharges],
        'TotalCharges': [TotalCharges]
    })

    X_processed = pipeline.transform(input_df)
    cat_cols = [col for col in input_df.columns if input_df[col].dtype == 'object']
    num_cols = [col for col in input_df.columns if input_df[col].dtype != 'object']

    cat_encoder = pipeline.named_steps['preprocessor'].named_transformers_['cat'].named_steps['onehot']
    cat_feature_names = cat_encoder.get_feature_names_out(cat_cols)

    feature_names = num_cols + list(cat_feature_names)
    final_df = pd.DataFrame(X_processed, columns=feature_names)
    final_df = pd.concat([final_df.iloc[:, 3:], final_df.iloc[:, :3]], axis=1)

    prediction_probs = model.predict_proba(final_df)[0]
    prediction_label = {
        "Prediction: CHURN 🔴": prediction_probs[1],
        "Prediction: STAY ✅": prediction_probs[0]
    }
    return prediction_label

# --- Personal Information Detection App ---
import gradio as gr
gr.load("models/iiiorg/piiranha-v1-detect-personal-information").launch()

# --- Gradio Interface ---
with gr.Blocks() as demo:
    gr.Markdown("# All-in-One AI Apps")
    with gr.Tab("Text Translator"):
        gr.HTML("<h1 align='center'>Text Translator</h1>")
        text_input = gr.Textbox(label="Enter Text")
        language_input = gr.Textbox(label="Enter Language")
        translate_btn = gr.Button("Translate")
        translated_text = gr.Textbox(label="Translated Text")
        translate_btn.click(fn=text_translator, inputs=[text_input, language_input], outputs=translated_text)

    with gr.Tab("Sentiment Analysis"):
        gr.Markdown("# Sentiment Analysis")
        sentiment_input = gr.Textbox(label="Enter Message")
        sentiment_output = gr.Textbox(label="Sentiment")
        sentiment_btn = gr.Button("Analyze Sentiment")
        sentiment_btn.click(fn=sentiment_analysis, inputs=sentiment_input, outputs=sentiment_output)

    with gr.Tab("Financial Analyst"):
        gr.Markdown("# Financial Analyst AI")
        financial_input = gr.Textbox(label="Enter Text for Financial Analysis")
        summarize_btn = gr.Button("Summarize")
        sentiment_btn = gr.Button("Classify Financial Tone")
        financial_output = gr.Textbox(label="Analysis Results")
        summarize_btn.click(fn=summarize_text, inputs=financial_input, outputs=financial_output)
        sentiment_btn.click(fn=text_to_sentiment, inputs=financial_input, outputs=financial_output)

    with gr.Tab("Customer Churn Prediction"):
        gr.Markdown("# Customer Churn Prediction")
        churn_inputs = [
            gr.Radio(['Yes', 'No'], label="Are you a Seniorcitizen?"),
            gr.Radio(['Yes', 'No'], label="Do you have a Partner?"),
            gr.Radio(['No', 'Yes'], label="Do you have Dependents?"),
            gr.Slider(label="Tenure (Months)", minimum=1, maximum=73),
            gr.Radio(['DSL', 'Fiber optic', 'No Internet'], label="Internet Service"),
            gr.Radio(['No', 'Yes'], label="Online Security"),
            gr.Radio(['No', 'Yes'], label="Online Backup"),
            gr.Radio(['No', 'Yes'], label="Device Protection"),
            gr.Radio(['No', 'Yes'], label="Tech Support"),
            gr.Radio(['No', 'Yes'], label="Streaming TV"),
            gr.Radio(['No', 'Yes'], label="Streaming Movies"),
            gr.Radio(['Month-to-month', 'One year', 'Two year'], label="Contract Type"),
            gr.Radio(['Yes', 'No'], label="Paperless Billing"),
            gr.Radio(['Electronic check', 'Mailed check', 'Bank transfer (automatic)', 'Credit card (automatic)'], label="Payment Method"),
            gr.Slider(label="Monthly Charges", minimum=18.4, maximum=118.65)
        ]
        churn_output = gr.Label(label="Churn Prediction")
        churn_btn = gr.Button("Predict Churn")
        churn_btn.click(fn=predict, inputs=churn_inputs, outputs=churn_output)

    with gr.Tab("Personal Information Detection"):
        gr.HTML("<h1 align='center'>Personal Information Detection</h1>")
        gr.Interface.load("models/iiiorg/piiranha-v1-detect-personal-information").launch()

demo.launch(share=True)