Nectar server note
server ip,
203.101.230.199 1
203.101.224.209 2
203.101.225.53 3
203.101.225.81 4
show hidden in file selection cmd+ shift +.
%ssh -Y -i nectar_shared_key_pair.pem ubuntu@203.101.230.199
ssh -Y ubuntu@203.101.230.199
老师安装了python3的tf，有些有问题，最终重新安装2.7下的
server 1 source /mnt/wanli/tensorflow_2/bin/activate
server 2 source /mnt/wanli/tensorflow_raw/bin/activate
server 3 source /mnt/wanli/tensorflow_2/bin_activate
/mnt/wanli
Make sure you have made these disks writable, or you will not be able to transfer data onto them. wanli and/or
python3 not working well
set new tensorflor using source /mnt/wanli/tensorflor_2/bin/active
nohup ./ shell > .log &
ps aux | grep wanli jobs
| grep ‘cpu ‘ /proc/stat | awk ‘{usage=($2+$4)*100/($2+$4+$5)} END {print usage “%”}’ |
sess = tf.InteractiveSession(config=tf.ConfigProto(intra_op_parallelism_threads=16))
config = tf.ConfigProto(intra_op_parallelism_threads=1, inter_op_parallelism_threads=1,
allow_soft_placement=True, device_count = {‘CPU’: 16})
session = tf.Session(config=config)