I have a pipeline where I sequentially run multiple segmentation models, but I only have a single GPU, and this process is running quite slowly. In my project, the number of models is even higher. Is there a way to speed up the execution of these models on a single GPU?
" Model 1 prediction
#pred_v01 = model_seg_v01.predict(image)
Some processing with pred_v01
Model 2 prediction
#pred_v02 = model_seg_v02.predict(image)
Some processing with pred_v02
Model 3 prediction
#pred_v03 = model_seg_v03.predict(image)
Some processing with pred_v03"
++ Platform Windows and Pytorch ++