Update app.py

app.py

@@ -14,6 +14,7 @@ bert_model.eval()
 model = tf.keras.models.load_model("rnn_Bi.h5")
 print("✅ Model loaded successfully!")
 
+# Function to extract BERT embedding
 def get_bert_embedding(text, max_length=100):
     inputs = tokenizer(text, return_tensors="pt", padding=True, truncation=True, max_length=max_length)
     with torch.no_grad():
@@ -23,16 +24,15 @@ def get_bert_embedding(text, max_length=100):
     embedding = embedding.reshape(1, 1, 768)  # shape (1, 1, 768)
     return embedding
 
+# Real sentiment prediction function using the model
 def predict_sentiment(text):
     embedding = get_bert_embedding(text)
     pred = model.predict(embedding)[0][0]
-    label = "Positive" if pred > 0.5 else "Negative"
+    label = "إيجابي" if pred > 0.5 else "سلبي"
     confidence = pred if pred > 0.5 else 1 - pred
-    return label, f"Confidence: {confidence:.2f}"
+    return label, f"{confidence * 100:.2f}%"
 
-import gradio as gr
-
-# Custom CSS for soft and pleasant colors like GPT style
+# Custom CSS for soft Arabic interface
 custom_css = """
 body {
     font-family: 'Segoe UI', Tahoma, Geneva, Verdana, sans-serif;
@@ -42,7 +42,7 @@ body {
     color: #202123;
 }
 .gradio-container h2, .gradio-container p {
-    color: #000000 !important;  /* لجعل عنوان ووصف بلون أسود */
+    color: #000000 !important;
 }
 .gradio-container {
     max-width: 600px;
@@ -53,7 +53,7 @@ body {
     box-shadow: 0 8px 24px rgba(32, 33, 35, 0.1);
 }
 .gr-button {
-    background-color: #4caf50 !important; /* soft green */
+    background-color: #4caf50 !important;
     color: white !important;
     font-weight: 600;
     border-radius: 12px !important;
@@ -89,39 +89,25 @@ body {
 .gr-textbox input[type="text"] {
     background-color: #f9fbfd !important;
 }
-
-/* تعديل لون التنبؤ داخل Label ليكون أسود */
 .gr-label .label-value,
 .gr-label .label-item,
 .gr-label .label,
 .gr-label span {
-    color: #FFFFFF !important;
+    color: #000000 !important;
 }
 """
 
-# Example sentiment prediction function
-def predict_sentiment(text):
-    # كودك الحقيقي هيكون هنا
-    if "جيد" in text or "رائع" in text or "ممتاز" in text:
-        return "إيجابي", "95%"
-    else:
-        return "سلبي", "82%"
-
-# Build Gradio interface with Blocks to add CSS
+# Build Gradio interface
 with gr.Blocks(css=custom_css) as iface:
     gr.Markdown("## تحليل المشاعر العربية بالذكاء الاصطناعي")
     gr.Markdown("اكتب جملة لتحليل المشاعر (إيجابي أو سلبي)")
 
     input_text = gr.Textbox(lines=2, placeholder="اكتب الجملة هنا...")
     sentiment_label = gr.Label(num_top_classes=2, label="المشاعر")
-    confidence_score = gr.Textbox(label="ثقة النموذج")
-
-    def wrapped_predict(text):
-        return predict_sentiment(text)
+    confidence_score = gr.Textbox(label="نسبة الثقة")
 
     btn = gr.Button("تحليل")
-    btn.click(fn=wrapped_predict, inputs=input_text, outputs=[sentiment_label, confidence_score])
+    btn.click(fn=predict_sentiment, inputs=input_text, outputs=[sentiment_label, confidence_score])
 
 if __name__ == "__main__":
     iface.launch()
-