把python函数转化为 tensorflow 函数 加速运算

docker run -tdi --gpus all  -p 60106:22 -p 60006:8888 --name  cujupt102 --privileged=true  registry.cn-hangzhou.aliyuncs.com/mkmk/gpujupyter:pydotGrapgviz /bin/bash && docker exec -d cujupt102 /bin/bash -c "cd /home &&(/etc/init.d/ssh start) && nohup jupyter notebook --allow-root & "

sshpass -p "Li" ssh -NTf -R 60006:172.16.102.168:60006 root@limengkai.work -o stricthostkeychecking=no

watch -n 1  nvidia-smi

pip install numpy sklearn matplotlib pandas tensorflow-gpu


sys.version_info(major=3, minor=7, micro=2, releaselevel='final', serial=0)
3.7.2 (v3.7.2:9a3ffc0492, Dec 24 2018, 02:44:43) 
[Clang 6.0 (clang-600.0.57)]
matplotlib 3.0.3
numpy 1.17.2
pandas 0.25.1
sklearn 0.21.3
tensorflow 2.0.0-beta0
tensorflow.python.keras.api._v2.keras 2.2.4-tf

import warnings 
warnings.filterwarnings('ignore')
import matplotlib as mpl
import matplotlib.pyplot as plt 
%matplotlib inline
import numpy as np
import sklearn 
import pandas as pd
import os
import sys
import time
import tensorflow as tf
from tensorflow import keras
from tensorflow.keras.layers import *

print(sys.version_info)
print(sys.version)
for module in mpl,np,pd,sklearn,tf,keras:
    print(module.__name__,module.__version__)
#tf.function  and auto-graph
def scaled_elu(z,scale=1.0,alpha=1.0):
    # z>=0? scale * z
    is_positive=tf.greater_equal(z,0.0)
    return scale * tf.where(is_positive,z,alpha*tf.nn.elu(z))

print(scaled_elu(tf.constant([-4.,-2])))

#根据python 方法转化成 tf 方法
scaled_elu_tf=tf.function(scaled_elu)
print(scaled_elu_tf(tf.constant([-4.,-3.])))

#根据 tf 方法 还原 python 方法
print(scaled_elu_tf.python_function is scaled_elu)

#tf 方法有什么优势，
#快!
%timeit scaled_elu(tf.random.normal((10000,10000)))
%timeit scaled_elu_tf(tf.random.normal((10000,10000)))
# 2.86 s ± 199 ms per loop (mean ± std. dev. of 7 runs, 1 loop each)
# 2.02 s ± 79.6 ms per loop (mean ± std. dev. of 7 runs, 1 loop each)
#这仅仅是在 cpu 测试的结果，如果采用 gpu 加速会更加明显