tensorflow-GPU usage guide

Personal public account

View the GPU usage of the machine

Regular update shows the condition of the gpu on the machine, refreshed once in #10s

nvidia-smi -l 10      

Dynamically apply for video memory

config = tf.ConfigProto()  
config.gpu_options.allow_growth = True  
session = tf.Session(config=config)  

Limit GPU usage

config = tf.ConfigProto()  
config.gpu_options.per_process_gpu_memory_fraction = 0.4  #占用40%显存  
session = tf.Session(config=config)  

Specify which GPU to use

CUDA_VISIBLE_DEVICES=1           Only device 1 will be seen
CUDA_VISIBLE_DEVICES=0,1         Devices 0 and 1 will be visible
CUDA_VISIBLE_DEVICES="0,1"       Same as above, quotation marks are optional
CUDA_VISIBLE_DEVICES=0,2,3       Devices 0, 2, 3 will be visible; device 1 is masked
CUDA_VISIBLE_DEVICES=""          No GPU will be visible

Set up in Python

os.environ['CUDA_VISIBLE_DEVICES'] = '0' #使用 GPU 0  
os.environ['CUDA_VISIBLE_DEVICES'] = '0,1' # 使用 GPU 0，1