Create app.py

app.py

@@ -0,0 +1,82 @@
+import gradio as gr
+from transformers import AutoTokenizer, AutoModelForCausalLM
+import torch
+import json
+
+# Load the ConspEmoLLM model
+MODEL_PATH = "lzw1008/ConspEmoLLM-v2"
+print("Loading ConspEmoLLM...")
+
+try:
+    tokenizer = AutoTokenizer.from_pretrained(MODEL_PATH)
+    model = AutoModelForCausalLM.from_pretrained(MODEL_PATH, device_map='auto')
+    print("Model loaded successfully!")
+    model_loaded = True
+except Exception as e:
+    print(f"Error: {e}")
+    tokenizer = None
+    model = None
+    model_loaded = False
+
+def analyze_text(text):
+    """Analyze text with ConspEmoLLM"""
+    if not model_loaded:
+        return {"error": "Model not loaded", "status": "error"}
+    
+    if not text.strip():
+        return {"error": "No text provided", "status": "error"}
+    
+    try:
+        inputs = tokenizer(text, return_tensors="pt", truncation=True, max_length=512)
+        
+        with torch.no_grad():
+            outputs = model.generate(
+                inputs["input_ids"], 
+                max_new_tokens=100,
+                temperature=0.7,
+                do_sample=True,
+                pad_token_id=tokenizer.eos_token_id
+            )
+        
+        result = tokenizer.batch_decode(outputs, skip_special_tokens=True)[0]
+        
+        return {
+            "result": result,
+            "original_text": text,
+            "status": "success"
+        }
+        
+    except Exception as e:
+        return {"error": str(e), "status": "error"}
+
+def api_interface(text):
+    """Main interface for API calls"""
+    if text.strip().lower() == "/health":
+        return json.dumps({"status": "healthy", "model_loaded": model_loaded}, indent=2)
+    
+    result = analyze_text(text)
+    return json.dumps(result, indent=2)
+
+# Create simple Gradio interface
+demo = gr.Interface(
+    fn=api_interface,
+    inputs=gr.Textbox(
+        lines=5, 
+        placeholder="Enter text to analyze...\nOr type '/health' to check status",
+        label="Input Text"
+    ),
+    outputs=gr.Textbox(
+        lines=10,
+        label="Analysis Result (JSON)"
+    ),
+    title="ConspEmoLLM API",
+    description="Analyze text for conspiracy theories and emotions",
+    examples=[
+        ["/health"],
+        ["I think the government is hiding vaccine information and this makes me worried."],
+        ["The media manipulates climate data to control us and I feel angry about it."]
+    ]
+)
+
+if __name__ == "__main__":
+    demo.launch()