Update app.py

app.py

@@ -11,34 +11,33 @@ from transformers import AutoTokenizer, AutoModelForSequenceClassification, Auto
 from huggingface_hub import hf_hub_download
 import spaces  # REQUIRED FOR ZEROGPU
 
-# ==========================================
+# ===================================
 # 1. LOAD AI MODELS (GLOBALLY CACHED)
-# ==========================================
+# ===================================
 print("Waking up the Genomic Oracle... Loading models into VRAM.")
-device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
 
 # A. Kadir's Gatekeeper (Local in the Space)
 clf_coding = joblib.load("coding_classifier_universal.joblib")
 
 # B. Base DNABERT (Local in the Space)
 tokenizer_base = AutoTokenizer.from_pretrained("DNABERT_Local", trust_remote_code=True)
-model_base = AutoModel.from_pretrained("DNABERT_Local", trust_remote_code=True).to(device)
+model_base = AutoModel.from_pretrained("DNABERT_Local", trust_remote_code=True)
 model_base.eval()
 
 # C. DNABERT-2 Promoter Model (Local in the Space)
 tokenizer_promoter = AutoTokenizer.from_pretrained("llm_promoter_classifier_v2", trust_remote_code=True)
-model_promoter = AutoModelForSequenceClassification.from_pretrained("llm_promoter_classifier_v2", trust_remote_code=True).to(device)
+model_promoter = AutoModelForSequenceClassification.from_pretrained("llm_promoter_classifier_v2", trust_remote_code=True)
 model_promoter.eval()
 
 # D. Multi-Feature LightGBM (Fetched from Cloud Repo)
-print("Downloading LightGBM from Model Repo...")
+print("Downloading LightGBM from Model Repository...")
 lgbm_path = hf_hub_download(repo_id="Geonomic/Genomic-Oracle-Weights", filename="dnabert_lightgbm_model_feature_type.pkl")
 lightgbm_model = joblib.load(lgbm_path)
 
 # E. Custom ALiBi Lean/Obese BERT (Fetched from Cloud Repo)
 print("Downloading Phenotype BERT from Model Repo...")
 tokenizer_pheno = AutoTokenizer.from_pretrained("Geonomic/Genomic-Oracle-Weights", trust_remote_code=True)
-model_pheno = AutoModelForSequenceClassification.from_pretrained("Geonomic/Genomic-Oracle-Weights", trust_remote_code=True).to(device)
+model_pheno = AutoModelForSequenceClassification.from_pretrained("Geonomic/Genomic-Oracle-Weights", trust_remote_code=True)
 model_pheno.eval()
 
 # Structural Feature Dictionary
@@ -47,11 +46,12 @@ FEATURE_DICT = {
     3: "3' UTR", 4: "Promoter", 5: "Enhancer", 6: "lncRNA"
 }
 
-# ==========================================
+# ==============================================
 # 2. CORE INFERENCE ENGINE (ZeroGPU Accelerated)
-# ==========================================
+# ==============================================
 @spaces.GPU  # This tells Hugging Face to teleport this specific math to the A100 GPU
 def run_deep_learning_cascade(dna_sequence):
+    device = torch.device("cuda")  # Wakes up the A100 connection
     clean_seq = "".join(dna_sequence.split()).upper()
     
     # --- LEVEL 1: Base Embedding & Kadir's Gatekeeper ---
@@ -109,9 +109,9 @@ def run_deep_learning_cascade(dna_sequence):
 
     return final_label, confidence, raw_scores
 
-# ==========================================
+# ===================================
 # 3. SPATIAL MAPPING (NCBI / ENSEMBL)
-# ==========================================
+# ===================================
 def get_genomic_context(sequence, is_coding):
     feature_type = "CODING" if is_coding else "PROMOTER"
     try:
@@ -159,9 +159,9 @@ def get_genomic_context(sequence, is_coding):
 
     return {"location": location_string, "start": start, "end": end, "strand": strand_txt, "metadata": gene_desc}
 
-# ==========================================
+# ==============================
 # 4. GRADIO INTERFACE (FRONTEND)
-# ==========================================
+# ==============================
 def gradio_inference(dna_sequence, run_mapping):
     if len(dna_sequence.strip()) < 10:
         return ("❌ Sequence too short!", "", "", "⚠️ Please enter at least 10 base pairs.")
@@ -187,16 +187,16 @@ def gradio_inference(dna_sequence, run_mapping):
             context_output = f"❌ Mapping failed: {context['error']}"
         elif "location" in context:
             context_lines = [
-                f"📍 **Location:** {context['location']}",
+                f"🎯 **Location:** {context['location']}",
                 f"🧬 **Strand:** {context['strand']}",
-                f"📏 **Coordinates:** {context['start']:,} – {context['end']:,}",
-                f"ℹ️ **Notes:** {context['metadata']}"
+                f"🧭 **Coordinates:** {context['start']:,} – {context['end']:,}",
+                f"🔬**Notes:** {context['metadata']}"
             ]
             context_output = "\n".join(context_lines)
         else:
             context_output = "⚠️ Could not map sequence."
     else:
-        context_output = "⏸️ Spatial mapping skipped (disable checkbox to run)."
+        context_output = "🗺️ Spatial mapping skipped (disable checkbox to run)."
 
     summary = (
         f"✅ Deep Scan Complete\n\n"
@@ -207,7 +207,7 @@ def gradio_inference(dna_sequence, run_mapping):
     return summary, "\n".join(stats_lines), context_output, ""
 
 # --- THE UI LAYOUT ---
-with gr.Blocks(theme=gr.themes.Soft(), title="🧬 The Genomic Oracle") as demo:
+with gr.Blocks(theme=gr.themes.Soft(), title="🧬 The Genomic Oracle 🧬") as demo:
     # The Integrated Landing Page
     gr.Markdown(
         """
@@ -234,7 +234,7 @@ with gr.Blocks(theme=gr.themes.Soft(), title="🧬 The Genomic Oracle") as demo:
         with gr.Column(scale=2):
             output_summary = gr.Textbox(label="✅ Classification Summary", lines=4)
             stats_panel = gr.Textbox(label="📊 Internal Pipeline Statistics", lines=4)
-            mapping_section = gr.Accordion("📍 Genomic Context (BLAST/Ensembl)", open=False)
+            mapping_section = gr.Accordion("📝 Genomic Context (BLAST/Ensembl)", open=False)
             with mapping_section:
                 context_output = gr.Textbox(label="Mapping Results", lines=5, placeholder="Results will appear here...")