Create app.py

app.py

@@ -0,0 +1,67 @@
+import gradio as gr
+from transformers import pipeline
+
+# Your private repo ID
+model_id = "Statistical-Impossibility/Feline-NER-Test"
+
+# Load NER pipeline
+ner_pipeline = pipeline("token-classification", model=model_id, aggregation_strategy="simple")
+
+def ner_predict(text):
+    results = ner_pipeline(text)
+    
+    # Sort by start position for clean order
+    results = sorted(results, key=lambda x: x['start'])
+    
+    # Build highlighted HTML
+    highlighted = ""
+    last_end = 0
+    for r in results:
+        # Add text before entity
+        highlighted += text[last_end:r['start']]
+        # Color map
+        color = {
+            "SYMPTOM": "yellow",
+            "DISEASE": "red",
+            "MEDICATION": "lightgreen",
+            "PROCEDURE": "lightblue",
+            "ANATOMY": "orange"
+        }.get(r['entity_group'], "gray")
+        # Highlight entity
+        highlighted += f"<mark style='background-color:{color}; padding:0.1em 0.2em; border-radius:3px;'>{r['word']}</mark>"
+        last_end = r['end']
+    highlighted += text[last_end:]  # Remaining text
+    
+    # Also return plain list
+    entity_list = "\n".join([f"{r['entity_group']}: {r['word']} (score: {r['score']:.3f})" for r in results])
+    
+    return highlighted, entity_list
+
+# Gradio interface
+with gr.Blocks(title="Feline Veterinary NER Demo") as demo:
+    gr.Markdown("# Feline Veterinary NER Demo")
+    gr.Markdown("Paste a veterinary text – model highlights entities.")
+    
+    with gr.Row():
+        with gr.Column(scale=2):
+            input_text = gr.Textbox(label="Input Text", lines=10, placeholder="Paste article/abstract here...")
+        with gr.Column(scale=2):
+            output_html = gr.HTML(label="Highlighted Output")
+            output_list = gr.Textbox(label="Extracted Entities", lines=10)
+    
+    btn = gr.Button("Run NER")
+    btn.click(
+        fn=ner_predict,
+        inputs=input_text,
+        outputs=[output_html, output_list]
+    )
+    
+    gr.Examples(
+        examples=[
+            ["A cat presented with vomiting and lethargy due to chronic kidney disease treated with subcutaneous fluids."],
+            ["Ultrasound revealed adrenal mass consistent with hyperaldosteronism."]
+        ],
+        inputs=input_text
+    )
+
+demo.launch()