Spaces:
Sleeping
Sleeping
| import gradio as gr | |
| from transformers import pipeline | |
| from langdetect import detect | |
| from huggingface_hub import InferenceClient | |
| import pandas as pd | |
| import os | |
| import nltk | |
| import asyncio | |
| nltk.download('punkt_tab') | |
| from nltk.tokenize import sent_tokenize | |
| HF_TOKEN = os.getenv("HF_TOKEN") | |
| # Fonction pour appeler l'API Zephyr avec des paramètres ajustés | |
| async def call_zephyr_api(prompt, mode, hf_token=HF_TOKEN): | |
| client = InferenceClient("HuggingFaceH4/zephyr-7b-beta", token=hf_token) | |
| try: | |
| if mode == "Rapide": | |
| max_new_tokens = 50 | |
| temperature = 0.3 | |
| elif mode == "Équilibré": | |
| max_new_tokens = 100 | |
| temperature = 0.5 | |
| else: # Précis | |
| max_new_tokens = 150 | |
| temperature = 0.7 | |
| response = await asyncio.to_thread(client.text_generation, prompt, max_new_tokens=max_new_tokens, temperature=temperature) | |
| return response | |
| except Exception as e: | |
| raise gr.Error(f"❌ Erreur d'appel API Hugging Face : {str(e)}") | |
| # Chargement du modèle de sentiment pour analyser les réponses | |
| classifier = pipeline("sentiment-analysis", model="mrm8488/distilroberta-finetuned-financial-news-sentiment-analysis") | |
| # Modèles de traduction (optionnels, désactivés pour optimisation) | |
| translator_to_en = pipeline("translation", model="Helsinki-NLP/opus-mt-mul-en") | |
| # Traduction en français désactivée pour l'instant | |
| # translator_to_fr = pipeline("translation", model="Helsinki-NLP/opus-mt-en-fr") | |
| # Fonction pour traduire un texte long en le segmentant (désactivée pour l'instant) | |
| def safe_translate_to_fr(text, max_length=512): | |
| return "Traduction désactivée pour l'instant pour améliorer la vitesse." | |
| # Fonction pour suggérer le meilleur modèle | |
| def suggest_model(text): | |
| word_count = len(text.split()) | |
| if word_count < 50: | |
| return "Rapide" | |
| elif word_count <= 200: | |
| return "Équilibré" | |
| else: | |
| return "Précis" | |
| # Fonction pour créer une jauge de sentiment | |
| def create_sentiment_gauge(sentiment, score): | |
| score_percentage = score * 100 | |
| if sentiment.lower() == "neutral": | |
| color = "#A9A9A9" | |
| elif sentiment.lower() == "positive": | |
| color = "#2E8B57" | |
| elif sentiment.lower() == "negative": | |
| color = "#DC143C" | |
| else: | |
| color = "#A9A9A9" | |
| html = f""" | |
| <div style='width: 100%; max-width: 300px; margin: 10px 0;'> | |
| <div style='background-color: #D3D3D3; border-radius: 5px; height: 20px; position: relative; box-shadow: 0 2px 4px rgba(0,0,0,0.2);'> | |
| <div style='background-color: {color}; width: {score_percentage}%; height: 100%; border-radius: 5px; transition: width 0.3s ease-in-out;'> | |
| </div> | |
| <span style='position: absolute; top: 0; left: 50%; transform: translateX(-50%); color: #0A1D37; font-size: 12px; line-height: 20px; font-weight: 600;'> | |
| {score_percentage:.1f}% | |
| </span> | |
| </div> | |
| <div style='text-align: center; font-size: 14px; margin-top: 5px; color: #E0E0E0;'> | |
| Sentiment: {sentiment} | |
| </div> | |
| </div> | |
| """ | |
| return html | |
| # Fonction d'analyse corrigée | |
| async def full_analysis(text, mode, detail_mode, count, history): | |
| if not text: | |
| yield "Entrez une phrase.", "", "", 0, history, None, "" | |
| return | |
| # Message de progression | |
| yield "Analyse en cours... (Étape 1 : Détection de la langue)", "", "", count, history, None, "" | |
| try: | |
| lang = detect(text) | |
| except: | |
| lang = "unknown" | |
| if lang != "en": | |
| text_en = translator_to_en(text, max_length=512)[0]['translation_text'] | |
| else: | |
| text_en = text | |
| yield "Analyse en cours... (Étape 2 : Analyse du sentiment)", "", "", count, history, None, "" | |
| # Analyse du sentiment avec RoBERTa sur le texte d'entrée | |
| result = await asyncio.to_thread(classifier, text_en) | |
| result = result[0] | |
| sentiment_output = f"Sentiment prédictif : {result['label']} (Score: {result['score']:.2f})" | |
| sentiment_gauge = create_sentiment_gauge(result['label'], result['score']) | |
| yield "Analyse en cours... (Étape 3 : Explication de l'impact)", "", "", count, history, None, "" | |
| # Appel à Zephyr pour expliquer l'impact basé sur le sentiment | |
| explanation_prompt = f"""<|system|> | |
| You are a professional financial analyst AI with expertise in economic forecasting. | |
| </s> | |
| <|user|> | |
| Given the following question about a potential economic event: "{text}" | |
| The predicted sentiment for this event is: {result['label'].lower()}. | |
| Assume the event happens (e.g., if the question is "Will the Federal Reserve raise interest rates?", assume they do raise rates). Explain why this event would likely have a {result['label'].lower()} economic impact. Provide a concise explanation in one paragraph, focusing on the potential effects on the economy. {"Use simple language for a general audience." if detail_mode == "Normal" else "Use detailed financial terminology for an expert audience."} | |
| </s> | |
| <|assistant|>""" | |
| explanation_en = await call_zephyr_api(explanation_prompt, mode) | |
| yield "Analyse en cours... (Étape 4 : Traduction)", "", "", count, history, None, "" | |
| # Traduction (désactivée pour l'instant) | |
| explanation_fr = safe_translate_to_fr(explanation_en, max_length=512) | |
| count += 1 | |
| history.append({ | |
| "Texte": text, | |
| "Sentiment": result['label'], | |
| "Score": f"{result['score']:.2f}", | |
| "Explication_EN": explanation_en, | |
| "Explication_FR": explanation_fr | |
| }) | |
| yield sentiment_output, explanation_en, explanation_fr, count, history, sentiment_gauge, "" | |
| # Fonction pour télécharger historique CSV | |
| def download_history(history): | |
| if not history: | |
| return None | |
| df = pd.DataFrame(history) | |
| file_path = "/tmp/analysis_history.csv" | |
| df.to_csv(file_path, index=False) | |
| return file_path | |
| # Interface Gradio améliorée | |
| def launch_app(): | |
| custom_css = """ | |
| @import url('https://fonts.googleapis.com/css2?family=Inter:wght@300;400;600;700&display=swap'); | |
| @import url('https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.0/css/all.min.css'); | |
| body { | |
| background-image: url('https://images.unsplash.com/photo-1593672715438-d88a70629abe?q=80&w=2070&auto=format&fit=crop'); | |
| background-size: cover; | |
| background-position: center; | |
| background-attachment: fixed; | |
| background-color: #0A1D37; | |
| background-blend-mode: overlay; | |
| color: #E0E0E0; | |
| font-family: 'Inter', sans-serif; | |
| margin: 0; | |
| padding: 20px; | |
| } | |
| .gr-box { | |
| background: #1A3C34 !important; | |
| border-radius: 12px !important; | |
| border: 1px solid #FFD700 !important; | |
| box-shadow: 0 6px 16px rgba(0, 0, 0, 0.4) !important; | |
| padding: 25px !important; | |
| margin: 15px 0 !important; | |
| transition: transform 0.2s ease, box-shadow 0.3s ease !important; | |
| } | |
| .gr-box:hover { | |
| transform: translateY(-3px) !important; | |
| box-shadow: 0 8px 20px rgba(255, 215, 0, 0.3) !important; | |
| } | |
| .gr-textbox, .gr-dropdown { | |
| background: #2E4A43 !important; | |
| border: 1px solid #FFD700 !important; | |
| border-radius: 8px !important; | |
| color: #E0E0E0 !important; | |
| font-size: 16px !important; | |
| padding: 12px !important; | |
| transition: border-color 0.3s ease, box-shadow 0.3s ease !important; | |
| } | |
| .gr-textbox:focus, .gr-dropdown:focus { | |
| border-color: #FFD700 !important; | |
| box-shadow: 0 0 10px rgba(255, 215, 0, 0.4) !important; | |
| } | |
| .gr-button { | |
| background: linear-gradient(90deg, #FFD700 0%, #D4AF37 100%) !important; | |
| color: #0A1D37 !important; | |
| border: none !important; | |
| border-radius: 8px !important; | |
| padding: 12px 24px !important; | |
| font-weight: 600 !important; | |
| font-size: 16px !important; | |
| transition: transform 0.1s ease, box-shadow 0.3s ease !important; | |
| box-shadow: 0 3px 10px rgba(255, 215, 0, 0.3) !important; | |
| } | |
| .gr-button:hover { | |
| transform: translateY(-2px) !important; | |
| box-shadow: 0 6px 14px rgba(255, 215, 0, 0.5) !important; | |
| } | |
| h1, h2, h3 { | |
| color: #FFD700 !important; | |
| font-weight: 700 !important; | |
| text-shadow: 0 2px 4px rgba(0, 0, 0, 0.3) !important; | |
| animation: fadeIn 1s ease-in-out; | |
| } | |
| @keyframes fadeIn { | |
| from { opacity: 0; transform: translateY(-10px); } | |
| to { opacity: 1; transform: translateY(0); } | |
| } | |
| .gr-row { | |
| margin: 20px 0 !important; | |
| } | |
| .gr-column { | |
| padding: 15px !important; | |
| } | |
| label { | |
| color: #FFD700 !important; | |
| font-weight: 600 !important; | |
| font-size: 16px !important; | |
| margin-bottom: 8px !important; | |
| display: flex !important; | |
| align-items: center !important; | |
| } | |
| label::before { | |
| font-family: "Font Awesome 6 Free"; | |
| font-weight: 900; | |
| margin-right: 8px; | |
| } | |
| .gr-textbox label::before { | |
| content: '\\f201'; | |
| } | |
| .gr-html label::before { | |
| content: '\\f080'; | |
| } | |
| .gr-file label::before { | |
| content: '\\f019'; | |
| } | |
| .economic-question-section { | |
| background: rgba(26, 60, 52, 0.9) !important; | |
| border-radius: 12px; | |
| padding: 25px; | |
| margin: 20px 0; | |
| box-shadow: 0 6px 16px rgba(0, 0, 0, 0.4); | |
| } | |
| .economic-question-section .gr-textbox { | |
| background: rgba(46, 74, 67, 0.8) !important; | |
| border: 2px solid #FFD700 !important; | |
| box-shadow: 0 3px 10px rgba(255, 215, 0, 0.3) !important; | |
| font-size: 18px !important; | |
| padding: 15px !important; | |
| } | |
| .economic-question-section .gr-textbox:focus { | |
| border-color: #FFD700 !important; | |
| box-shadow: 0 0 12px rgba(255, 215, 0, 0.5) !important; | |
| } | |
| .prompt-box { | |
| background: rgba(46, 74, 67, 0.6) !important; | |
| border: 1px solid #FFD700 !important; | |
| border-radius: 8px !important; | |
| color: #E0E0E0 !important; | |
| font-size: 16px !important; | |
| padding: 12px !important; | |
| margin-top: 10px !important; | |
| } | |
| .prompt-box label::before { | |
| content: '\\f075'; | |
| } | |
| .options-section { | |
| display: flex; | |
| flex-direction: column; | |
| gap: 15px; | |
| margin-top: 15px; | |
| } | |
| .options-section .gr-dropdown { | |
| width: 200px !important; | |
| } | |
| .options-section .gr-dropdown label::before { | |
| content: '\\f0c9'; | |
| } | |
| .progress-message { | |
| color: #FFD700 !important; | |
| font-style: italic; | |
| margin-bottom: 10px; | |
| } | |
| """ | |
| with gr.Blocks(theme=gr.themes.Base(), css=custom_css) as iface: | |
| gr.Markdown("# 📈 Analyse Financière Premium + Explication IA", elem_id="title") | |
| gr.Markdown("Posez une question sur un événement économique. L'IA analyse le sentiment et explique l'impact.", elem_classes=["subtitle"]) | |
| count = gr.State(0) | |
| history = gr.State([]) | |
| with gr.Row(elem_classes=["economic-question-section"]): | |
| with gr.Column(scale=2): | |
| with gr.Column(): | |
| input_text = gr.Textbox( | |
| lines=4, | |
| label="Question Économique" | |
| ) | |
| prompt_display = gr.Textbox( | |
| value="Une hausse des taux causerait-elle une récession ?", | |
| label="Exemple de Prompt", | |
| interactive=False, | |
| elem_classes=["prompt-box"] | |
| ) | |
| with gr.Column(scale=1, elem_classes=["options-section"]): | |
| mode_selector = gr.Dropdown( | |
| choices=["Rapide", "Équilibré", "Précis"], | |
| value="Équilibré", | |
| label="Mode (longueur et style de réponse)" | |
| ) | |
| detail_mode_selector = gr.Dropdown( | |
| choices=["Normal", "Expert"], | |
| value="Normal", | |
| label="Niveau de détail (simplicité ou technicité)" | |
| ) | |
| with gr.Row(): | |
| analyze_btn = gr.Button("Analyser") | |
| reset_graph_btn = gr.Button("Réinitialiser") | |
| download_btn = gr.Button("Télécharger CSV") | |
| with gr.Row(): | |
| with gr.Column(scale=1): | |
| progress_message = gr.Textbox(label="Progression", elem_classes=["progress-message"], interactive=False) | |
| sentiment_output = gr.Textbox(label="Sentiment Prédictif") | |
| sentiment_gauge = gr.HTML(label="Jauge de Sentiment") | |
| with gr.Column(scale=2): | |
| with gr.Row(): | |
| explanation_output_en = gr.Textbox(label="Explication en Anglais") | |
| explanation_output_fr = gr.Textbox(label="Explication en Français") | |
| download_file = gr.File(label="Fichier CSV") | |
| input_text.change(lambda t: gr.update(value=suggest_model(t)), inputs=[input_text], outputs=[mode_selector]) | |
| analyze_btn.click( | |
| full_analysis, | |
| inputs=[input_text, mode_selector, detail_mode_selector, count, history], | |
| outputs=[sentiment_output, explanation_output_en, explanation_output_fr, count, history, sentiment_gauge, progress_message] | |
| ) | |
| download_btn.click( | |
| download_history, | |
| inputs=[history], | |
| outputs=[download_file] | |
| ) | |
| iface.launch(share=True) | |
| if __name__ == "__main__": | |
| launch_app() |