Mask_RCNN icon indicating copy to clipboard operation
Mask_RCNN copied to clipboard

Published 20 hours ago • verified publisher icon matterport

how to inference with cpu only

Open Pengfeicc opened this issue 2 years ago • 1 comments

i have installed tensorflow with cpu version and keras and other environments successfully, pip install tensorflow==1.15.0 pip install keras==2.2.4

but how to set a config when i want to inference with cpu? For example in visualize.py, should i set GPU_COUNT = 0 ? but here are a comment that Batch size = GPU_COUNT * IMAGES_PER_GPU, then Batch size will be 0 ???? i am newbie please help me

Pengfeicc avatar Apr 21 '22 01:04 Pengfeicc

i did not install the tensorflow-gpu in my conda environment, but when i set GPU_COUNT = 1, it seems to use my gpu to inference, why??? Screenshot from 2022-04-21 04-07-27

Pengfeicc avatar Apr 21 '22 02:04 Pengfeicc