文章目录:
一.TensorFlow简介
二.TF安装流程
三.TF学习路线
四.基础案例
五.总结
https://github.com/eastmountyxz/
AI-for-TensorFlow
https://github.com/eastmountyxz/
AI-for-Keras
学Python近八年,认识了很多大佬和朋友,感恩。由于在外求学且需要养娃,故在CSDN设置成了最低价收费专栏,觉得不错的可以购买抬爱;但作者的本意是帮助更多初学者入门,因此在github开源了所有代码,也在公众号同步更新。深知自己很菜,得拼命努力前行,编程也没有什么捷径,干就对了。希望未来能更透彻学习和撰写文章,也能在读博几年里学会真正的独立科研。同时非常感谢参考文献中的大佬们的文章和分享,共勉。
- https://blog.csdn.net/eastmount
pip 安装
virtualenv安装
docker安装
从安装源安装
Linux \ MacOS \ Windows
CPU版 \ GPU版(GPU版本比CPU版本快很多倍)
测试版
更新TensorFlow
Windows10
CPU
TensorFlow2.0
Anaconda
Python3.6
//检查Anaconda是否成功安装
conda --version
//检测目前安装了哪些环境
conda info --envs
conda search --full-name python
tensorflow2.0GPU版本的环境配置与安装教程 normalization
[Tensorflow2.0] Tensorflow2.0的安装教程 - 牛andmore牛
conda create -n tf2 python=3.6
activate tf2
pip install tensorflow==2.0.0-alpha0
conda search tensorflow #搜CPU版
conda search tensorflow-gpu #搜GPU版
conda install tensorflow=2.0.0 #安装CPU版
conda install tensorflow-gpu=2.0.0 #安装GPU版
import tensorflow as tf
print(tf.__version__)
2.环境配置
4.TensorFlow
核心知识点如下,推荐读者学习,作者也分享。
TensorFlow基础知识
Tensor
Session
Variable
Placeholder
Dropout
Tensorboard
CNN
RNN(GRU\LSTM)
Autoencoder
GNN
神经网络参数保存
神经网络性能评价
入门课程推荐一:CS231n(斯坦福大学 李飞飞教授)
入门课程推荐二:CS224n(斯坦福大学 Christopher Manning教授)
入门课程推荐三:《Deep Learning》(斯坦福大学 吴恩达大佬)
入门课程推荐四:CS224w 图神经网络(斯坦福大学 Jurij Leskovec大佬)
入门课程推荐五:李宏毅《机器学习》(B站视频)
# -*- coding: utf-8 -*-
"""
Spyder Editor
This is a temporary script file.
By:Eastmount CSDN YXZ 2019-11-28
"""
import tensorflow as tf
#查询TensorFlow版本
print(tf.__version__)
#定义a和b为两个常量
a = tf.constant([1, 2], name="a")
b = tf.constant([2, 3], name="b")
print(a)
print(b)
#随机生成一个正态分布
output = tf.random.normal([5,3])
print(output)
#创建2个矩阵并进行相乘
matrix1 = tf.constant([[3,3]])
matrix2 = tf.constant([[1, 2],[3, 4]])
product = tf.matmul(matrix1,matrix2)
print(matrix1)
print(matrix2)
print(product)
print(product.numpy())
2.0.0-alpha0
tf.Tensor([1 2], shape=(2,), dtype=int32)
tf.Tensor([2 3], shape=(2,), dtype=int32)
tf.Tensor(
[[-2.1826832 -0.32986134 -1.6238695 ]
[-0.18214056 0.25923613 -0.12570491]
[ 1.0550841 -0.6655764 -1.5837296 ]
[-0.10004017 0.0162886 0.9483853 ]
[ 0.4709251 -0.18713968 0.8347026 ]], shape=(5, 3), dtype=float32)
tf.Tensor([[3 3]], shape=(1, 2), dtype=int32)
tf.Tensor(
[[1 2]
[3 4]], shape=(2, 2), dtype=int32)
tf.Tensor([[12 18]], shape=(1, 2), dtype=int32)
[[12 18]]
天行健,君子以自强不息。
地势坤,君子以厚德载物。
(By: Eastmount 2021-06-18 夜于武汉)
参考文献如下,感谢这些大佬!
神经网络和机器学习基础入门分享 - Eastmount
Stanford机器学习—第五讲. 神经网络的学习 Neural Networks learning
吴祖增前辈:神经网络入门(连载之一)
吴祖增前辈:神经网络入门(连载之二)
斯坦福机器学习视频NG教授:
https://class.coursera.org/ml/class/index
网易云莫烦老师视频(强推)
https://study.163.com/course/courseLearn.htm?
courseId=1003209007
https://zhuanlan.zhihu.com/p/148312957 - DrugAI好友