Update app.py

app.py

@@ -16,7 +16,7 @@ def call_zephyr_api(prompt, hf_token=HF_TOKEN):
     except Exception as e:
         raise gr.Error(f"❌ Erreur d'appel API Hugging Face : {str(e)}")
 
-# Chargement du modèle de sentiment
+# Chargement du modèle de sentiment pour analyser les réponses
 classifier = pipeline("sentiment-analysis", model="mrm8488/distilroberta-finetuned-financial-news-sentiment-analysis")
 
 # Modèles de traduction
@@ -33,10 +33,38 @@ def suggest_model(text):
     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 = "gray"
+    elif sentiment.lower() == "positive":
+        color = "green"
+    elif sentiment.lower() == "negative":
+        color = "red"
+    else:
+        color = "gray"
+
+    html = f"""
+    <div style='width: 100%; max-width: 300px; margin: 10px 0;'>
+        <div style='background-color: #e0e0e0; border-radius: 5px; height: 20px; position: relative;'>
+            <div style='background-color: {color}; width: {score_percentage}%; height: 100%; border-radius: 5px;'>
+            </div>
+            <span style='position: absolute; top: 0; left: 50%; transform: translateX(-50%); color: black; font-size: 12px; line-height: 20px;'>
+                {score_percentage:.1f}%
+            </span>
+        </div>
+        <div style='text-align: center; font-size: 14px; margin-top: 5px;'>
+            Sentiment: {sentiment}
+        </div>
+    </div>
+    """
+    return html
+
 # Fonction d'analyse
 def full_analysis(text, mode, detail_mode, count, history):
     if not text:
-        return "Entrez une phrase.", "", "", 0, history, None
+        return "Entrez une phrase.", "", "", 0, history, None, ""
 
     try:
         lang = detect(text)
@@ -46,26 +74,38 @@ def full_analysis(text, mode, detail_mode, count, history):
     if lang != "en":
         text = translator_to_en(text, max_length=512)[0]['translation_text']
 
-    result = classifier(text)[0]
-    sentiment_output = f"Sentiment : {result['label']} (Score: {result['score']:.2f})"
+    # Étape 1 : Poser une question à Zephyr pour prédire l'impact économique
+    prediction_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}"
+
+Assume the event happens (e.g., if the question is "Will the Federal Reserve raise interest rates?", assume they do raise rates). What would be the likely economic impact of this event? Provide a concise explanation in one paragraph, focusing on the potential positive or negative effects on the economy. Do not repeat the question or the prompt in your response.
+</s>
+<|assistant|>"""
+    prediction_response = call_zephyr_api(prediction_prompt)
+
+    # Étape 2 : Analyser le sentiment de la réponse de Zephyr
+    result = classifier(prediction_response)[0]
+    sentiment_output = f"Sentiment prédictif : {result['label']} (Score: {result['score']:.2f})"
+    sentiment_gauge = create_sentiment_gauge(result['label'], result['score'])
 
-    prompt = f"""<|system|>
+    # Étape 3 : Générer une explication détaillée
+    explanation_prompt = f"""<|system|>
 You are a professional financial analyst AI.
 </s>
 <|user|>
-Analyze the following financial news carefully:
-"{text}"
+Given the following question about a potential economic event: "{text}"
 
-The detected sentiment for this news is: {result['label'].lower()}.
+Based on your prediction of the economic impact, which is: "{prediction_response}"
 
-Now, explain why the sentiment is {result['label'].lower()} using a logical, fact-based explanation.
-Base your reasoning only on the given news text.
-Do not repeat the news text or the prompt.
-Respond only with your financial analysis in one clear paragraph.
-Write in a clear and professional tone.
+The predicted sentiment for this impact is: {result['label'].lower()}.
+
+Now, explain why the sentiment is {result['label'].lower()} using a logical, fact-based explanation. Base your reasoning only on the predicted economic impact. Respond only with your financial analysis in one clear paragraph. Write in a clear and professional tone.
 </s>
 <|assistant|>"""
-    explanation_en = call_zephyr_api(prompt)
+    explanation_en = call_zephyr_api(explanation_prompt)
     explanation_fr = translator_to_fr(explanation_en, max_length=512)[0]['translation_text']
 
     count += 1
@@ -77,7 +117,7 @@ Write in a clear and professional tone.
         "Explication_FR": explanation_fr
     })
 
-    return sentiment_output, explanation_en, explanation_fr, count, history
+    return sentiment_output, explanation_en, explanation_fr, count, history, sentiment_gauge
 
 # Fonction pour télécharger historique CSV
 def download_history(history):
@@ -92,13 +132,13 @@ def download_history(history):
 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.")
+        gr.Markdown("Entrez une question sur un événement économique. L'IA prédit l'impact et attribue un sentiment (positif, négatif, neutre).")
 
         count = gr.State(0)
         history = gr.State([])
 
         with gr.Row():
-            input_text = gr.Textbox(lines=4, placeholder="Entrez une actualité ici...", label="Texte à analyser")
+            input_text = gr.Textbox(lines=4, placeholder="Entrez une question ici (ex. 'La Réserve fédérale augmentera-t-elle ses taux d'intérêt avant 2025 ?')", label="Question économique")
 
         with gr.Row():
             mode_selector = gr.Dropdown(
@@ -117,7 +157,9 @@ def launch_app():
         download_btn = gr.Button("Télécharger CSV")
 
         with gr.Row():
-            sentiment_output = gr.Textbox(label="Résultat du Sentiment")
+            sentiment_output = gr.Textbox(label="Résultat du Sentiment Prédictif")
+
+        sentiment_gauge = gr.HTML(label="Jauge de Sentiment")
 
         with gr.Row():
             with gr.Column():
@@ -132,7 +174,7 @@ def launch_app():
         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]
+            outputs=[sentiment_output, explanation_output_en, explanation_output_fr, count, history, sentiment_gauge]
         )
 
         download_btn.click(