chore: download onnx-data on spaces

Signed-off-by: Suvaditya Mukherjee <suvadityamuk@gmail.com>

app.py

@@ -102,7 +102,9 @@ if __name__ == "__main__":
 
     ONNX_MODEL_PATH = "https://huggingface.co/onnx-community/Qwen2.5-1.5B-Instruct/resolve/main/onnx/model.onnx_data"
 
+    print("Downloading ONNX model...")
     response = requests.get(ONNX_MODEL_PATH, stream=True)
+    print("Downloaded ONNX model!")
 
     # Download file
     download_pdf_from_gdrive(RESUME_URL, RESUME_PATH)
@@ -149,9 +151,9 @@ if __name__ == "__main__":
     #     }
     # )
 
-    wandb.login(
-        key=os.getenv("WANDB_API_KEY")
-    )
+    # wandb.login(
+    #     key=os.getenv("WANDB_API_KEY")
+    # )
 
     @spaces.GPU
     def rag_process(message, chat_history):

utils.py

@@ -181,7 +181,7 @@ def scrape_website(start_url, delay=1):
 
 def onnx_inference(chat_history, rag_query, tokenizer):
     # Create ONNX Runtime session
-    session = onnxruntime.InferenceSession("model.onnx", providers=['CUDAExecutionProvider', 'CPUExecutionProvider'])
+    session = onnxruntime.InferenceSession("model.onnx")
     
     # Tokenize input text
     inputs = tokenizer.apply_chat_template(