[BUG]: Memory leak when run Yolo example

[Shamaseen-iotistic]

Bug description

I ran the YOLO example from "max/examples/inference/yolo-python-onnx" and it worked well. However, I noticed that memory usage kept increasing until the device crashed.

To diagnose the issue, I used Memray to trace the program. My findings suggest that the memory leak originates from the api.py file.

how to solve it ?

System information

AMD® Ryzen threadripper 2990wx 32-core processor × 64
Ubuntu 22.04.5 LTS
magic 0.3.0
mojo 24.5.0 (e8aacb95)
max 24.5.0 (e8aacb95)
Modular version 24.5.0