深度学习算法与编程

转载：

https://blog.csdn.net/oBrightLamp/article/details/85067981#RNN_138

文章目录

        前言
            本书内容
            资料推荐
            开源许可 LICENSE
            软件版本
        损失函数
            MSELoss
            cross-entropy
            softmax
            softmax + cross-entropy
        优化算法
            正则化 / 参数规范惩罚
            SGD, Momentum, RMSprop, Adam
            Batch Normalization
        特征工程
            PCA
        全连接神经网络
            Affine
        卷积神经网络
            convolution
            Transpose Convolution
            MaxPool
            ReLU
            dropout
        循环神经网络
            RNN
            LSTM
            GRU
        生成对抗网络
            GAN
        实战
            Kaggle
        附录

前言
本书内容

    内容提要
    – https://blog.csdn.net/oBrightLamp/article/details/85162129

资料推荐

    自学深度学习之计算机视觉的入门资料推荐
    – https://blog.csdn.net/oBrightLamp/article/details/84076410

开源许可 LICENSE

所有的说明性文档基于 Creative Commons 协议, 所有的代码基于 MIT 协议.
All documents are licensed under the Creative Commons License, all codes are licensed under the MIT License.
软件版本

Python = 3.6
scikit-learn = 0.20.0
TensorFlow = 1.12
PyTorch = 1.0
损失函数
MSELoss

    均方差损失函数MSELoss详解及反向传播中的梯度求导
    – https://blog.csdn.net/oBrightLamp/article/details/85137756

cross-entropy

    通过函数图像介绍信息熵的概念
    – https://blog.csdn.net/oBrightLamp/article/details/85269091

    案例详解cross-entropy交叉熵损失函数及反向传播
    – https://blog.csdn.net/oBrightLamp/article/details/83962147

    Python和PyTorch对比实现cross-entropy交叉熵损失函数及反向传播
    – https://blog.csdn.net/oBrightLamp/article/details/84029058

softmax

    softmax函数详解及误差反向传播的梯度求导
    – https://blog.csdn.net/oBrightLamp/article/details/83959185

    纯Python和PyTorch对比实现softmax及其反向传播
    – https://blog.csdn.net/oBrightLamp/article/details/84034658

softmax + cross-entropy

    多标签softmax + cross-entropy交叉熵损失函数详解及反向传播中的梯度求导
    – https://blog.csdn.net/oBrightLamp/article/details/84069835

    Python和PyTorch对比实现多标签softmax + cross-entropy交叉熵损失及反向传播
    – https://blog.csdn.net/oBrightLamp/article/details/84073485

优化算法
正则化 / 参数规范惩罚

    L2正则化Regularization详解及反向传播的梯度求导
    – https://blog.csdn.net/oBrightLamp/article/details/85290929

SGD, Momentum, RMSprop, Adam

    常用梯度下降算法SGD, Momentum, RMSprop, Adam详解
    – https://blog.csdn.net/oBrightLamp/article/details/85218783

    纯Python和PyTorch对比实现SGD, Momentum, RMSprop, Adam梯度下降算法
    – https://blog.csdn.net/oBrightLamp/article/details/85218799

Batch Normalization

    Batch Normalization函数详解及反向传播中的梯度求导
    – https://blog.csdn.net/oBrightLamp/article/details/84332455

    Python和PyTorch对比实现批标准化Batch Normalization函数及反向传播
    – https://blog.csdn.net/oBrightLamp/article/details/84557854

    Batch Normalization的测试或推理过程及样本参数更新方法
    – https://blog.csdn.net/oBrightLamp/article/details/85391056

    Python和PyTorch对比实现批标准化 Batch Normalization 函数在测试或推理过程中的算法
    – https://blog.csdn.net/oBrightLamp/article/details/85391167

特征工程
PCA

    特征工程PCA降维方法的最大方差理论详解
    – https://blog.csdn.net/oBrightLamp/article/details/85255895

    纯Python和scikit-learn对比实现PCA特征降维
    – https://blog.csdn.net/oBrightLamp/article/details/85255898

全连接神经网络
Affine

    affine/linear(仿射/线性)变换函数详解及全连接层反向传播的梯度求导
    – https://blog.csdn.net/oBrightLamp/article/details/84333111

    Python和PyTorch对比实现affine/linear(仿射/线性)变换函数及全连接层的反向传播
    – https://blog.csdn.net/oBrightLamp/article/details/84453996

卷积神经网络
convolution

    卷积convolution函数详解及反向传播中的梯度求导
    – https://blog.csdn.net/oBrightLamp/article/details/84561088

    Python和PyTorch对比实现卷积convolution函数及反向传播
    – https://blog.csdn.net/oBrightLamp/article/details/84589545

    卷积convolution函数的矩阵化计算方法及其梯度的反向传播
    – https://blog.csdn.net/oBrightLamp/article/details/85870773

    Python 实现 TensorFlow 和 PyTorch 验证卷积 convolution 函数矩阵化计算及反向传播
    – https://blog.csdn.net/oBrightLamp/article/details/85870813

Transpose Convolution

    TensorFlow和PyTorch对比理解卷积和反向卷积或转置卷积(Transpose Convolution)
    – https://blog.csdn.net/oBrightLamp/article/details/85708124

MaxPool

    池化层MaxPool函数详解及反向传播的公式推导
    – https://blog.csdn.net/oBrightLamp/article/details/84635346

    Python和PyTorch对比实现池化层MaxPool函数及反向传播
    – https://blog.csdn.net/oBrightLamp/article/details/84635308

ReLU

    ReLU函数详解及反向传播中的梯度求导
    – https://blog.csdn.net/oBrightLamp/article/details/84326978

    Python和PyTorch对比实现ReLU函数及反向传播
    – https://blog.csdn.net/oBrightLamp/article/details/84326804

dropout

    dropout函数详解及反向传播中的梯度求导
    – https://blog.csdn.net/oBrightLamp/article/details/84105097

    Python和PyTorch对比实现dropout函数及反向传播
    – https://blog.csdn.net/oBrightLamp/article/details/84326091

循环神经网络
RNN

    循环神经网络RNNCell单元详解及反向传播的梯度求导
    – https://blog.csdn.net/oBrightLamp/article/details/85015325

    纯Python和PyTorch对比实现循环神经网络RNNCell及反向传播
    – https://blog.csdn.net/oBrightLamp/article/details/85015402

    纯Python和PyTorch对比实现循环神经网络RNN及反向传播
    – https://blog.csdn.net/oBrightLamp/article/details/85015387

LSTM

    长短期记忆网络LSTMCell单元详解及反向传播的梯度求导
    – https://blog.csdn.net/oBrightLamp/article/details/85068285

    纯Python和PyTorch对比实现循环神经网络LSTM及反向传播
    – https://blog.csdn.net/oBrightLamp/article/details/85069255

GRU

    门控循环单元GRUCell详解及反向传播的梯度求导
    – https://blog.csdn.net/oBrightLamp/article/details/85109589

    纯Python和PyTorch对比实现门控循环单元GRU及反向传播
    – https://blog.csdn.net/oBrightLamp/article/details/85109607

生成对抗网络
GAN

    生成对抗网络 GAN 的数学原理
    – https://blog.csdn.net/oBrightLamp/article/details/86553074

实战
Kaggle

    PyTorch Kaggle 快速上手(杂草幼苗图片识别)
    – https://blog.csdn.net/oBrightLamp/article/details/84947499

附录

    常用Latex公式
    – https://blog.csdn.net/oBrightLamp/article/details/83964331
---------------------
作者：BrightLampCsdn
来源：CSDN
原文：https://blog.csdn.net/oBrightLamp/article/details/85067981
版权声明：本文为博主原创文章，转载请附上博文链接！