Update app.py

app.py

@@ -12,26 +12,24 @@ load_dotenv()
 genai.configure(api_key=os.getenv("GEMINI_API_KEY"))
 model_gemini = genai.GenerativeModel("gemini-pro")
 
-def get_gemini_advice(emotion):
-    prompt = f"""Give me one verse from the Qur'an or Hadith that is relevant to someone feeling {emotion}.
-Format it like this:
+def get_gemini_advice_from_text(text):
+    prompt = f"""Suggest one Qur'an verse or Hadith to comfort and guide a person who wrote the following:
 
-Arabic:
-[Verse Arabic]
+'{text}'
 
-English Translation:
-[Translation]
+Please include:
+- Arabic verse or hadith
+- English translation
+- Source (Surah or Hadith reference)
 
-Source:
-[Surah or Hadith Reference]"""
+Wrap it in a gentle tone."""
     try:
         response = model_gemini.generate_content(prompt)
-        return f"""<div class='notion-card animated'>
-<h3>{emotion.capitalize()}</h3>
+        return f"""<div class='notion-card fade-in'>
 <p style='white-space: pre-wrap;'>{response.text}</p>
 </div>"""
     except Exception as e:
-        return f"<div class='notion-card'><h3>{emotion.capitalize()}</h3><p>May Allah guide your heart. (AI fallback failed)</p></div>"
+        return f"<div class='notion-card fade-in'><p>May Allah guide your heart.</p></div>"
 
 model_name = "SamLowe/roberta-base-go_emotions"
 tokenizer = AutoTokenizer.from_pretrained(model_name)
@@ -80,24 +78,54 @@ def analyze_combined(text, threshold, image):
         img_label = None
 
     final_label = text_label if img_label is None else img_label
-    card = islamic_advice.get(final_label) or get_gemini_advice(final_label)
+    card = islamic_advice.get(final_label)
+    if not card:
+        card = get_gemini_advice_from_text(text)
     return card
 
-def analyze_batch_csv(file):
-    df = pd.read_csv(file.name)
-    results = []
-    for text in df['text']:
-        inputs = tokenizer(text, return_tensors="pt", truncation=True, padding=True)
-        with torch.no_grad():
-            logits = model(**inputs).logits
-        probs = sigmoid(logits)[0]
-        idx = torch.argmax(probs).item()
-        label = model.config.id2label[idx].lower()
-        advice = label.capitalize()
-        results.append({"text": text, "emotion": label, "advice": advice})
-    return pd.DataFrame(results)
-
 custom_css = """
+@keyframes slideInUp {
+  from {
+    transform: translateY(20px);
+    opacity: 0;
+  }
+  to {
+    transform: translateY(0);
+    opacity: 1;
+  }
+}
+@keyframes glowFadeIn {
+  0% {
+    box-shadow: 0 0 0px rgba(0, 0, 0, 0);
+    opacity: 0;
+  }
+  100% {
+    box-shadow: 0 0 15px rgba(59, 130, 246, 0.3);
+    opacity: 1;
+  }
+}
+.fade-in {
+  animation: fadeInPop 0.6s ease-out both;
+}
+.slide-up {
+  animation: slideInUp 0.6s ease-out both;
+}
+.glow-card {
+  animation: glowFadeIn 1s ease-in-out both;
+}
+@keyframes fadeInPop {
+  0% {
+    opacity: 0;
+    transform: scale(0.95);
+  }
+  100% {
+    opacity: 1;
+    transform: scale(1);
+  }
+}
+.fade-in {
+  animation: fadeInPop 0.6s ease-out both;
+}
 body {
   background: #f9fafb;
   font-family: 'Inter', sans-serif;
@@ -129,9 +157,5 @@ with gr.Blocks(css=custom_css) as demo:
         result = gr.HTML()
         btn.click(fn=analyze_combined, inputs=[text_input, threshold_slider, image_input], outputs=result)
 
-    with gr.Tab("Batch Analysis"):
-        file_input = gr.File(file_types=[".csv"], label="Upload CSV with 'text' column")
-        table_output = gr.Dataframe()
-        file_input.change(fn=analyze_batch_csv, inputs=file_input, outputs=table_output)
-
+    
 demo.launch()