Ubuntu-tensorflow program end GPU Memory not released issue

I ran TensorFlow program on Ubuntu, halfway through the use of the Win+c key to the end of the program, but the GPU video memory is not released, has been in the occupied state.
Using commands
Nvidia-smi
Shown below

Two GPU programs are in progress, in fact, gpu:0 has been stopped by the author, but the GPU is not released, the process is still continuing, so only the use of violence, the process manually shut down, process number in the red line, because the author in two GPU run program, it is difficult to find themselves from the program name, But can find their own PID from the Gpu:num.
The close command is as follows:
sudo kill-9 PID

Good, perfect solution to GPU memory release issues.

Ubuntu-tensorflow program end GPU Memory not released issue