@kpatrick
2019-11-14T19:25:18.000000Z
字数 1577
阅读 66
work
vivo
daily
ssh 11111198@vivoadmin@172.25.71.133@passsh.vivo.xyz -p 22
ssh 11111198@passsh.vivo.xyz -p 22
后跳转: vivoadmin
172.25.71.133
scp -r local_dir 11111198@vivoadmin@172.25.71.133@passsh.vivo.xyz:/data04/cxj
prefix
): conda env export > environment.yaml
conda env create -f environment.yml
conda list -e > conda_list.txt
conda create -n new_environment --file conda_list.txt
pip freeze > requirements.txt
pip install -r requirements.txt
--------- 防踩坑记录 ----------
nvidis-smi
查看驱动和cuda
NVIDIA-SMI 384.183 Driver Version: 384.183 CUDA Version: 9.0
查看conda环境下的cudnn和cuda版本是否与显卡驱动匹配
conda安装tensorflow-gpu时,版本超过1.10.0之后,cuda版本为9.2,cudnn为7.6.4
$ conda list | grep -E 'tensorflow|cuda|cudnn'
WARNING: The conda.compat module is deprecated and will be removed in a future release.
cudatoolkit 9.2 0
cudnn 7.6.4 cuda9.2_0
tensorflow 1.12.0 gpu_py36he74679b_0
tensorflow-base 1.12.0 gpu_py36had579c0_0
tensorflow-gpu 1.12.0 h0d30ee6_0
查看官方的推荐方案
tensorflow版本和cuda、cudnn版本:1.5.0到1.12.0和cuda 9兼容,
conda安装:安装tensorflow-gpu的时候,超过1.10.0时,conda方案安装的版本是9.2,会导致兼容性问题
用文韬给的脚本,提取一个wav文件的filterbank和mfcc特征,后续继续补充知识。
import os
# gpu visible
os.environ["CUDA_VISIBLE_DEVICES"]= '0'
# ignore tensorflow information
os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'
0 = all messages are logged (default behavior)
1 = INFO messages are not printed
2 = INFO and WARNING messages are not printed
3 = INFO, WARNING, and ERROR messages are not printed