anaconda3-5.2.0
官网
https://www.anaconda.com/
历史版本下载地址
1 2 3 https://repo.continuum.io/archive/ https://repo.anaconda.com/archive/ https://mirrors.tuna.tsinghua.edu.cn/anaconda/archive/?C=N&O=D
win10安装
https://blog.51cto.com/acevi/2103437
linux 7 安装 anaconda 3-5.2.0 tensorflow1.11.0
https://blog.51cto.com/moerjinrong/2155178
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 安装 chmod +x Anaconda3-5.2.0-Linux-x86_64.sh ./Anaconda3-5.2.0-Linux-x86_64.sh 安装过程中会需要不断回车来阅读并同意license。安装路径默认为用户目录(可以自己指定),最后需要确认将路径加入用户的.bashrc中。 In order to continue the installation process, please review the license agreement. Please, press ENTER to continue >>> # 要继续安装过程,请查看许可证协议。请按ENTER继续 然后按空格阅读许可协议, Do you accept the license terms? [yes|no] [no] >>> yes # 是否接受协议,选yes Anaconda3 will now be installed into this location: /root/anaconda3 - Press ENTER to confirm the location - Press CTRL-C to abort the installation - Or specify a different location below [/root/anaconda3] >>> # 是否安装到当前家目录的anaconda3目录中,默认回车即可 Do you wish the installer to prepend the Anaconda3 install location to PATH in your /root/.bashrc ? [yes|no] [no] >>> yes # 是否添加环境变量到/root/.bashrc文件 重新加载环境变量,执行: source ~/.bashrc python -V pip list conda list
silent install
1 [xy_zhangpeng@bigdata-2 ~]$ bash Anaconda3-5.2.0-Linux-x86_64.sh -b -p $HOME/anaconda3 -f
To run the silent installation of Miniconda for macOS or Linux, specify the -b and -p arguments of the bash installer. The following arguments are supported:
-b—Batch mode with no PATH modifications to ~/.bashrc. Assumes that you agree to the license agreement. Does not edit the .bashrc or .bash_profile files.
-p—Installation prefix/path.
-f—Force installation even if prefix -p already exists.
tensorflow 在 anaconda。。
https://www.anaconda.com/tensorflow-in-anaconda/
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 安装Tensorflow 如果GPU是NVIDIA的,就可以安装GPU版本的TensorFlow;如果不是,安装CPU版本的就好了。 1.因为要下载Tensorflow,所以我先在Anaconda的配置文件中添加清华镜像库,这样下载和更新的速度会快很多,命令: conda config --add channels https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/free/ conda config --set show_channel_urls yes #查看 conda config --show #CPU版本 pip install --upgrade tensorflow #GPU版本 pip install --upgrade tensorflow-gpu #指定版本 pip install tensorflow==1.11.0
分布式TF
https://www.jianshu.com/p/fdb93e44a8cc
例子
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 # 引入 tensorflow 模块 import tensorflow as tf # 创建一个整型常量,即 0 阶 Tensor t0 = tf.constant(3, dtype=tf.int32) # 创建一个浮点数的一维数组,即 1 阶 Tensor t1 = tf.constant([3., 4.1, 5.2], dtype=tf.float32) # 创建一个字符串的2x2数组,即 2 阶 Tensor t2 = tf.constant([['Apple', 'Orange'], ['Potato', 'Tomato']], dtype=tf.string) # 创建一个 2x3x1 数组,即 3 阶张量,数据类型默认为整型 t3 = tf.constant([[[5], [6], [7]], [[4], [3], [2]]]) # 打印上面创建的几个 Tensor print(t0) print(t1) print(t2) print(t3) --------------------- 作者:戈云飞 来源:CSDN 原文:https://blog.csdn.net/geyunfei_/article/details/78782804 版权声明:本文为博主原创文章,转载请附上博文链接!
参考
https://cloud.tencent.com/developer/article/1078485
https://cloud.tencent.com/developer/article/1078028
parcel包安装
1 2 3 https://repo.anaconda.com/pkgs/misc/parcels/ 替换为 https://repo.continuum.io/pkgs/misc/parcels/
离线下载tensorlfow。 Keras
方法:在有网络的机器下载好。zip到离线服务器。
1 2 3 4 5 6 7 pip download msgpack-python==0.5.6 -d pip_package mkdir pip_package pip download tensorflow==1.11.0 -d pip_package zip -r pip_package.zip pip_package
1 2 3 4 5 6 7 8 9 10 http://mirrors.aliyun.com/pypi/simple/tensorflow/ http://mirrors.aliyun.com/pypi/packages/2c/0c/74410a32bf753b280b28b685dc6620c65ccc3a09494398d47198af9f2bbb/tensorflow-1.11.0rc2-cp36-cp36m-manylinux1_x86_64.whl#sha256=b137211744ccbfec6fd5a5f62a47ce1a467fd760be8169a38c7a88121e8f6341 http://mirrors.aliyun.com/pypi/packages/5e/10/aa32dad071ce52b5502266b5c659451cfd6ffcbf14e6c8c4f16c0ff5aaab/Keras-2.2.4-py2.py3-none-any.whl#sha256=794d0c92c6c4122f1f0fcf3a7bc2f49054c6a54ddbef8d8ffafca62795d760b6 pip install xxx.whl
anaconda cdh parcels
1 2 3 4 https://www.cloudera.com/downloads/partner/anaconda.html https://docs.anaconda.com/anaconda-scale/cloudera-cdh/ http://docs.anaconda.com/anaconda-repository/user-guide/tasks/work-with-cloudera-parcels/ https://www.anaconda.com/how-to-generate-custom-anaconda-parcels-for-cloudera-cdh-with-anaconda-enterprise-5/
histroy
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 1063 bash Anaconda3-5.2.0-Linux-x86_64.sh -b -p /opt/anaconda3 -f 1064 ll /opt/ 1065 python -V 1066 ll 1067 cd /opt/ 1068 ll 1069 cd anaconda3/ 1070 ll 1071 ./bin/python -V 1072 ./bin/conda -V 1074 ./bin/conda config -h 1075 ./bin/conda config --show 1077 ./bin/conda config --add channels https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/free/ 1078 ./bin/conda config --show 1080 ./bin/conda config --set show_channel_urls yes 1081 ./bin/conda config --show 1082 ./pip -V 1083 ./bin/pip -V 1084 ./bin/pip install tensorflow==1.11.0 1085 ./bin/pip list all
Tensorflow报错 tf.estimator package not installed
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 1、tf.estimator package 未安装 tf.estimator package not installed 我使用的环境是: anaconda 5.2.0, Python 3.6和TensorFlow 1.12 然后在网上找了一下解决方案之后,发现需要更新一下numpy、pandas和matplotlib等package: pip install -U pandas pip install -U matplotlib 将pandas的版本更新到了0.23.4,matplotlib更新到3.0.2就好了,然后import TensorFlow的时候就不会报错了 2、import tensorflow的提醒 FutureWarning: Conversion of the second argument of issubdtype from `float` to `np.floating` is deprecated. In future, it will be treated as `np.float64 == np.dtype(float).type`. from ._conv import register_converters as _register_converters 这个问题还是环境问题,h5py出了问题,对h5py进行一个升级了就,后面再import tensorflow的时候,就不会出错了。 pip install h5py==2.8.0rc1 3、 TypeError: __init__() got an unexpected keyword argument 'serialized_options' 很有可能是,终端上的 protoc 版本 与python库内的protobuf版本不一样。 pip install -U protobuf 4、将Python程序不挂断的跑到服务器上面, 注意CUDA_VISIBLE_DEVICES=0需要放在nohup的前面。 CUDA_VISIBLE_DEVICES=0 nohup python -u main.py 作者:奔向算法的喵 链接:https://www.jianshu.com/p/2ea51363b080 来源:简书 简书著作权归作者所有,任何形式的转载都请联系作者获得授权并注明出处。