Update app.py

app.py

@@ -1,173 +1,132 @@
 import gradio as gr
 import requests
-from transformers import pipeline
-from langdetect import detect
 import pandas as pd
 import textstat
-import matplotlib.pyplot as plt
 import os
+from transformers import pipeline
 
+# Récupération du token Hugging Face
 HF_TOKEN = os.getenv("HF_TOKEN")
 
-
-# Fonction pour appeler l'API Mistral-7B
+# Fonction pour appeler l'API Zephyr-7B-Beta
 def call_zephyr_api(prompt, hf_token=HF_TOKEN):
     API_URL = "https://api-inference.huggingface.co/models/HuggingFaceH4/zephyr-7b-beta"
     headers = {"Authorization": f"Bearer {hf_token}"}
     payload = {"inputs": prompt, "parameters": {"max_new_tokens": 300}}
-    
+
     try:
         response = requests.post(API_URL, headers=headers, json=payload, timeout=60)
         response.raise_for_status()
-        return response.json()[0]["generated_text"]
+        return response.json()[0]["generated_text"].strip()
     except Exception as e:
-        raise gr.Error(f"❌ Erreur d'appel API Hugging Face : {str(e)}")
+        raise gr.Error(f"Erreur d'appel API Hugging Face : {str(e)}")
+
+# Pipeline d'analyse de sentiment initial (optionnel)
+classifier = pipeline("sentiment-analysis", model="mrm8488/distilroberta-finetuned-financial-news-sentiment-analysis")
 
+# Fonction principale d'analyse
+def full_analysis(text, history):
+    if not text:
+        return "Entrez une phrase.", "", 0, history, None
 
+    # 1. Demander à Zephyr de donner uniquement le sentiment
+    prompt_sentiment = f"""
+You are a financial news sentiment detector.
 
+Given the following news text:
+"{text}"
 
-# Chargement du modèle de sentiment
-classifier = pipeline("sentiment-analysis", model="mrm8488/distilroberta-finetuned-financial-news-sentiment-analysis")
+Respond only with one word: positive, neutral, or negative.
 
-# Modèles de traduction
-translator_to_en = pipeline("translation", model="Helsinki-NLP/opus-mt-mul-en")
-translator_to_fr = pipeline("translation", model="Helsinki-NLP/opus-mt-en-fr")
-
-# 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 générer un graphique de clarté
-def plot_clarity(clarity_scores):
-    plt.figure(figsize=(8, 4))
-    plt.plot(range(1, len(clarity_scores) + 1), clarity_scores, marker='o')
-    plt.title("Évolution du Score de Clarté")
-    plt.xlabel("Numéro d'analyse")
-    plt.ylabel("Score de Clarté")
-    plt.ylim(0, 100)
-    plt.grid(True)
-    return plt.gcf()
-
-# Fonction pour reset le graphique
-def reset_clarity_graph():
-    return [], plot_clarity([])
-
-# Fonction d'analyse
-def full_analysis(text, mode, detail_mode, count, history, clarity_scores):
-    if not text:
-        return "Entrez une phrase.", "", "", 0, history, clarity_scores, None, None
+Do not add any explanation or extra text.
+"""
+    detected_sentiment = call_zephyr_api(prompt_sentiment).lower()
 
-    try:
-        lang = detect(text)
-    except:
-        lang = "unknown"
+    if detected_sentiment not in ["positive", "neutral", "negative"]:
+        detected_sentiment = "neutral"
 
-    if lang != "en":
-        text = translator_to_en(text, max_length=512)[0]['translation_text']
+    # 2. Demander l'explication basée sur ce sentiment
+    prompt_explanation = f"""
+You are a financial analyst AI.
 
-    result = classifier(text)[0]
-    sentiment_output = f"Sentiment : {result['label']} (Score: {result['score']:.2f})"
+Given the following financial news:
+"{text}"
 
-    prompt = f"""
-    You are a financial analyst AI.
-    Based on the following financial news: \"{text}\",
-    explain clearly why the sentiment is {result['label'].lower()}.
-    {"Write a concise paragraph." if detail_mode == "Normal" else "Write a detailed explanation over multiple paragraphs."}
-    """
+The detected sentiment is: {detected_sentiment}.
 
-    explanation_en = call_zephyr_api(prompt)
-    explanation_fr = translator_to_fr(explanation_en, max_length=512)[0]['translation_text']
+Now explain clearly why the sentiment is {detected_sentiment}.
+Write a concise paragraph.
+"""
+    explanation = call_zephyr_api(prompt_explanation)
 
-    clarity_score = textstat.flesch_reading_ease(explanation_en)
-    clarity_scores.append(clarity_score)
+    # 3. Calculer le score de clarté
+    clarity_score = textstat.flesch_reading_ease(explanation)
+    clarity_score = max(0, min(clarity_score, 100))  # Bornage 0-100
 
-    count += 1
+    # 4. Stocker dans l'historique
     history.append({
         "Texte": text,
-        "Sentiment": result['label'],
-        "Score": f"{result['score']:.2f}",
-        "Explication_EN": explanation_en,
-        "Explication_FR": explanation_fr,
-        "Clarté": f"{clarity_score:.1f}"
+        "Sentiment": detected_sentiment.capitalize(),
+        "Clarté": f"{clarity_score:.1f}",
+        "Explication": explanation
     })
 
-    return sentiment_output, explanation_en, explanation_fr, clarity_score, count, history, clarity_scores, plot_clarity(clarity_scores)
-
-# Fonction pour télécharger historique CSV
+    return detected_sentiment.capitalize(), explanation, clarity_score, history, clarity_score
+
+# Fonction pour générer la barre de clarté
+def generate_clarity_bar(score, sentiment):
+    color = "green" if sentiment.lower() == "positive" else ("red" if sentiment.lower() == "negative" else "gray")
+    return gr.BarPlot.update(
+        value=[["Clarity", score]],
+        x="label",
+        y="value",
+        colors=[color],
+        width=400,
+        height=50,
+        y_lim=[0, 100]
+    )
+
+# Fonction pour télécharger l'historique
 def download_history(history):
     if not history:
         return None
     df = pd.DataFrame(history)
-    file_path = "/tmp/analysis_history.csv"
+    file_path = "/tmp/history.csv"
     df.to_csv(file_path, index=False)
     return file_path
 
-# Interface Gradio
+# Gradio Interface
 def launch_app():
-    with gr.Blocks(theme=gr.themes.Base(), css="body {background-color: #0D1117; color: white;} .gr-button {background-color: #161B22; border: 1px solid #30363D;}") as iface:
-
-        gr.Markdown("# 📈 Analyse Financière Premium + Explication IA", elem_id="title")
-        gr.Markdown("Entrez une actualité financière. L'IA analyse et explique en anglais/français. Choisissez votre mode d'explication.")
-
-        count = gr.State(0)
-        history = gr.State([])
-        clarity_scores = gr.State([])
+    with gr.Blocks() as iface:
+        gr.Markdown("# 📈 Analyse Financière Premium - Zephyr7B")
 
         with gr.Row():
-            input_text = gr.Textbox(lines=4, placeholder="Entrez une actualité ici...", label="Texte à analyser")
+            input_text = gr.Textbox(label="Entrez votre question financière", lines=3)
 
         with gr.Row():
-            mode_selector = gr.Dropdown(
-                choices=["Rapide", "Équilibré", "Précis"],
-                value="Équilibré",
-                label="Mode recommandé selon la taille"
-            )
-            detail_mode_selector = gr.Dropdown(
-                choices=["Normal", "Expert"],
-                value="Normal",
-                label="Niveau de détail"
-            )
-
-        analyze_btn = gr.Button("Analyser")
-        reset_graph_btn = gr.Button("Reset Graphique")
-        download_btn = gr.Button("Télécharger CSV")
+            analyze_btn = gr.Button("Analyser")
+            download_btn = gr.Button("Télécharger l'historique")
 
         with gr.Row():
-            sentiment_output = gr.Textbox(label="Résultat du Sentiment")
+            sentiment_output = gr.Textbox(label="Sentiment Détecté")
+            clarity_bar = gr.BarPlot()
 
-        with gr.Row():
-            with gr.Column():
-                explanation_output_en = gr.Textbox(label="Explication en Anglais")
-            with gr.Column():
-                explanation_output_fr = gr.Textbox(label="Explication en Français")
+        explanation_output = gr.Textbox(label="Explication de l'IA", lines=5)
+        clarity_score_text = gr.Textbox(label="Score de Clarté (%)")
+        file_output = gr.File(label="Fichier CSV")
 
-        clarity_score_output = gr.Textbox(label="Score de Clarté (Flesch Reading Ease)")
-        clarity_plot = gr.Plot(label="Graphique des Scores de Clarté")
-        download_file = gr.File(label="Fichier CSV")
-
-        input_text.change(lambda t: gr.update(value=suggest_model(t)), inputs=[input_text], outputs=[mode_selector])
+        history = gr.State([])
 
         analyze_btn.click(
             full_analysis,
-            inputs=[input_text, mode_selector, detail_mode_selector, count, history, clarity_scores],
-            outputs=[sentiment_output, explanation_output_en, explanation_output_fr, clarity_score_output, count, history, clarity_scores, clarity_plot]
-        )
-
-        reset_graph_btn.click(
-            reset_clarity_graph,
-            outputs=[clarity_scores, clarity_plot]
+            inputs=[input_text, history],
+            outputs=[sentiment_output, explanation_output, clarity_score_text, history, clarity_bar]
         )
 
         download_btn.click(
             download_history,
             inputs=[history],
-            outputs=[download_file]
+            outputs=[file_output]
         )
 
     iface.launch()