Changed the device configuration to solve issues

app.py

@@ -24,7 +24,7 @@ try:
     )
     model = AutoPeftModelForCausalLM.from_pretrained(
         MODEL_NAME,
-        device_map="auto",  # Will use CPU if GPU not available
+        device_map=None,  # Will use CPU if GPU not available
         trust_remote_code=True,
         cache_dir = '/app/cache'
         #load_in_4bit=True

test.py

@@ -0,0 +1,37 @@
+import requests
+
+# Replace with your actual Space URL
+API_URL = "https://sidharthan-scriptr.hf.space/generate"
+
+def test_single_request():
+    # Test data
+    test_prompt = """
+    <bos><start_of_turn>user
+    how are you?
+    <start_of_turn>model
+    """
+    
+    # Make request
+    try:
+        response = requests.post(
+            API_URL,
+            json={
+                "message": test_prompt,
+                "temperature": 0.7,
+                "max_length": 512
+            }
+        )
+        
+        # Print results
+        print(f"Status Code: {response.status_code}")
+        if response.status_code == 200:
+            print("\nGenerated Text:")
+            print(response.json()['generated_text'])
+        else:
+            print(f"Error: {response.text}")
+            
+    except Exception as e:
+        print(f"Error making request: {str(e)}")
+
+if __name__ == "__main__":
+    test_single_request()