STATUS_ENTRYPOINT_NOT_FOUND error under tch 0.18.0

[Siflorite]

I copied the code from examples/basics.rs and tried to run it on Windows. I have installed CUDA 12.4 on my computer and have set the environments of libtorch. I've downloaded the libtorch 2.5.0 cu124 for windows at https://download.pytorch.org/libtorch/cu124/libtorch-win-shared-with-deps-2.5.0%2Bcu124.zip according to the readme file. The program can be successfully complied when I'm using dependency

[dependencies]
tch = {version = "0.18.0"}

, but it throws the error

error: process didn't exit successfully: `target\debug\yolo_self.exe` (exit code: 0xc0000139, STATUS_ENTRYPOINT_NOT_FOUND)

when I tried to execute cargo run.
Then I tried libtorch 2.5.0 under cu121, cu118 and cpu and they all failed to run under the same error. When I switched to using tch = "0.17.0" and libtorch 2.4.0 cu124 downloaded from https://download.pytorch.org/libtorch/cu124/libtorch-win-shared-with-deps-2.4.0%2Bcu124.zip, the program can be successfully compiled and run. So I think that it might be the problem that tch 0.18.0 cannot support libtorch 2.5.0.