# This cell will generate a unified Gradio app.py content based on all 5 apps provided import os import gradio as gr import pandas as pd import numpy as np import joblib import spacy from transformers import pipeline from langchain_core.pydantic import BaseModel, Field from langchain.prompts import HumanMessagePromptTemplate, ChatPromptTemplate from langchain.output_parsers import PydanticOutputParser from langchain_openai import ChatOpenAI # ---------------- Text Translator ---------------- # chat = ChatOpenAI() class TextTranslator(BaseModel): output: str = Field(description="Translated output text") output_parser = PydanticOutputParser(pydantic_object=TextTranslator) format_instructions = output_parser.get_format_instructions() def text_translator(input_text: str, language: str) -> str: human_template = f"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) return output.output # ---------------- Sentiment Analysis ---------------- # sentiment_classifier = pipeline("sentiment-analysis", model="cardiffnlp/twitter-xlm-roberta-base-sentiment") def sentiment_analysis(message, history): result = sentiment_classifier(message) return f"Sentiment: {result[0]['label']} (Probability: {result[0]['score']:.2f})" # ---------------- Financial Analyst ---------------- # 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 = [res['label'] for res in results] return list(zip(split_in_sentences(text), results_list)) auth_token = os.environ.get("HF_Token") asr = pipeline("automatic-speech-recognition", "facebook/wav2vec2-base-960h") def speech_to_text(speech): return asr(speech)["text"] summarizer = pipeline("summarization", model="knkarthick/MEETING_SUMMARY") def summarize_text(text): return summarizer(text)[0]['summary_text'] fin_model = pipeline("sentiment-analysis", model='yiyanghkust/finbert-tone', tokenizer='yiyanghkust/finbert-tone') def text_to_sentiment(text): return fin_model(text)[0]["label"] def fin_ner(text): api = gr.Interface.load("dslim/bert-base-NER", src='models', use_auth_token=auth_token) return api(text) def fin_ext(text): results = fin_model(split_in_sentences(text)) return make_spans(text, results) def fls(text): fls_model = pipeline("text-classification", model="demo-org/finbert_fls", tokenizer="demo-org/finbert_fls", use_auth_token=auth_token) results = fls_model(split_in_sentences(text)) return make_spans(text, results) # ---------------- Personal Information Identifier ---------------- # def detect_personal_info(text): pii_model = gr.Interface.load("models/iiiorg/piiranha-v1-detect-personal-information") return pii_model(text) # ---------------- Customer Churn ---------------- # 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_churn = joblib.load(pipeline_path) model_churn = joblib.load(model_path) def calculate_total_charges(tenure, monthly_charges): return tenure * monthly_charges def predict_churn(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] }) 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'] X_processed = pipeline_churn.transform(input_df) cat_encoder = pipeline_churn.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) first_three_columns = final_df.iloc[:, :3] remaining_columns = final_df.iloc[:, 3:] final_df = pd.concat([remaining_columns, first_three_columns], axis=1) prediction_probs = model_churn.predict_proba(final_df)[0] return { "Prediction: CHURN 🔴": prediction_probs[1], "Prediction: STAY ✅": prediction_probs[0] } # ---------------- Interface ---------------- # with gr.Blocks() as app: with gr.Tab("Text Translator"): input_text = gr.Textbox(label="Enter text to translate") lang = gr.Textbox(label="Target language (e.g., Hindi, French)") output_text = gr.Textbox(label="Translated text") gr.Button("Translate").click(fn=text_translator, inputs=[input_text, lang], outputs=output_text) with gr.Tab("Sentiment Analysis"): gr.ChatInterface(sentiment_analysis) with gr.Tab("Financial Analyst"): audio_input = gr.Audio(source="microphone", type="filepath") text = gr.Textbox(label="Transcribed Text") gr.Button("Transcribe").click(fn=speech_to_text, inputs=audio_input, outputs=text) stext = gr.Textbox(label="Summary") gr.Button("Summarize").click(fn=summarize_text, inputs=text, outputs=stext) gr.Button("Financial Tone").click(fn=text_to_sentiment, inputs=stext, outputs=gr.Label()) gr.Button("NER").click(fn=fin_ner, inputs=text, outputs=gr.HighlightedText()) gr.Button("Tone per sentence").click(fn=fin_ext, inputs=text, outputs=gr.HighlightedText()) gr.Button("Forward-looking").click(fn=fls, inputs=text, outputs=gr.HighlightedText()) with gr.Tab("Personal Information Identifier"): pii_input = gr.Textbox(label="Enter text to analyze") pii_output = gr.Textbox(label="Detected Personal Info") gr.Button("Detect").click(fn=detect_personal_info, inputs=pii_input, outputs=pii_output) with gr.Tab("Customer Churn"): churn_inputs = [ gr.Radio(['Yes', 'No'], label="SeniorCitizen"), gr.Radio(['Yes', 'No'], label="Partner"), gr.Radio(['No', 'Yes'], label="Dependents"), gr.Slider(1, 73, step=1, label="Tenure (Months)"), gr.Radio(['DSL', 'Fiber optic', 'No Internet'], label="InternetService"), gr.Radio(['No', 'Yes'], label="OnlineSecurity"), gr.Radio(['No', 'Yes'], label="OnlineBackup"), gr.Radio(['No', 'Yes'], label="DeviceProtection"), gr.Radio(['No', 'Yes'], label="TechSupport"), gr.Radio(['No', 'Yes'], label="StreamingTV"), gr.Radio(['No', 'Yes'], label="StreamingMovies"), gr.Radio(['Month-to-month', 'One year', 'Two year'], label="Contract"), gr.Radio(['Yes', 'No'], label="PaperlessBilling"), gr.Radio(['Electronic check', 'Mailed check', 'Bank transfer (automatic)', 'Credit card (automatic)'], label="PaymentMethod"), gr.Slider(18.40, 118.65, label="MonthlyCharges") ] churn_output = gr.Label(label="Churn Prediction") gr.Button("Predict").click(fn=predict_churn, inputs=churn_inputs, outputs=churn_output) app.launch()