2018年05月_小锋子Shawn

原创「Deep Learning」Note on SFT-GAN（空间特征变换GAN）

Sina Weibo：小锋子Shawn Tencent E-mail：403568338@qq.com http://blog.csdn.net/dgyuanshaofeng/article/details/80494787涉及超分辨率重建中的纹理恢复。[1] Recovering Realistic Texture in Image Super-resolution by Dee...

2018-05-31 20:26:19 1637

原创「Deep Learning」Note on Super-FAN（超分人脸对齐网络）

Sina Weibo：小锋子Shawn Tencent E-mail：403568338@qq.com http://blog.csdn.net/dgyuanshaofeng/article/details/80494787[1] Super-FAN Integrated Facial Landmark Localization and Super-resolution of Real...

2018-05-31 09:01:46 2013

原创「Machine Learning」Note on Neural Network and Random Forest（神经网络与随机森林）

Sina Weibo：小锋子Shawn Tencent E-mail：403568338@qq.com http://blog.csdn.net/dgyuanshaofeng/article/details/80379520深度学习与随机森林的结合，有如下算法： deep regression forest neural regression forest deep neural...

2018-05-21 11:46:07 3158 1

原创「Deep Learning」Note on MaskLab

Sina Weibo：小锋子Shawn Tencent E-mail：403568338@qq.com http://blog.csdn.net/dgyuanshaofeng/article/details/80370826[1] MaskLab Instance Segmentation by Refining Object Detection with Semantic and D...

2018-05-20 01:31:34 1508

原创「Deep Learning」Note on Recurrent Salienct Transformation Network

Sina Weibo：小锋子Shawn Tencent E-mail：403568338@qq.com http://blog.csdn.net/dgyuanshaofeng/article/details/80370826[1] Recurrent Saliency Transformation Network Incorporating Multi-Stage Visual Cue...

2018-05-20 01:25:27 509

原创「Deep Learning」Note on MSDNet（多尺度密集连接网络）

Sina Weibo：小锋子Shawn Tencent E-mail：403568338@qq.com http://blog.csdn.net/dgyuanshaofeng/article/details/80370826ICLR2018 Oral论文[1] Multi-Scale Dense Networks for Resource Efficient Image Class...

2018-05-19 12:19:49 1321

原创「Deep Learning」Note on NADM

Sina Weibo：小锋子Shawn Tencent E-mail：403568338@qq.com http://blog.csdn.net/dgyuanshaofeng/article/details/80370826[1] Incorporating Nesterov Momentum into Adam 2016 [paper]

2018-05-19 00:42:54 119

原创「Deep Learning」Note on ADADELTA

Sina Weibo：小锋子Shawn Tencent E-mail：403568338@qq.com http://blog.csdn.net/dgyuanshaofeng/article/details/80370826[1] Adadelta An Adaptive Learning Rate Method 2012 [paper]

2018-05-19 00:39:53 114

原创「Deep Learning」Note on Adam

Sina Weibo：小锋子Shawn Tencent E-mail：403568338@qq.com http://blog.csdn.net/dgyuanshaofeng/article/details/80370826[1] Adam A Method for Stochastic Optimization 2014 [paper]

2018-05-19 00:37:44 459

原创「Deep Learning」Note on RMSprop

Sina Weibo：小锋子Shawn Tencent E-mail：403568338@qq.com http://blog.csdn.net/dgyuanshaofeng/article/details/80370826[1] RMSprop: Divide the gradient by a running average of its recent magnitude 2012...

2018-05-19 00:35:58 233

原创「Deep Learning」Note on ADAGrad（比vanilla SGD好的优化算法）

Sina Weibo：小锋子Shawn Tencent E-mail：403568338@qq.com http://blog.csdn.net/dgyuanshaofeng/article/details/80370826[1] Adaptive Subgradient Methods for Online Learning and Stochastic Optimization 2...

2018-05-19 00:27:19 463

原创「Deep Learning」Note on AMSGrad（比Adam好的优化算法）

Sina Weibo：小锋子Shawn Tencent E-mail：403568338@qq.com http://blog.csdn.net/dgyuanshaofeng/article/details/80209602[1] On the Convergence of Adam and Beyond 2018 [paper]

2018-05-18 23:56:19 2031

原创「Pillow/PIL」Note on Image Module

Sina Weibo：小锋子Shawn Tencent E-mail：403568338@qq.com http://blog.csdn.net/dgyuanshaofeng/article/details/78764882基于 Pillow 5.0.0，文档 Image Module，文档Image.convert，对图像进行转换，mode有[L, RGB, CMYK]，但是mat...

2018-05-17 02:41:16 123

原创「Deep Learning」理解Pytorch中的「torch.nn」

Sina Weibo：小锋子Shawn Tencent E-mail：403568338@qq.com http://blog.csdn.net/dgyuanshaofeng/article/details/78768107Module，所有神经网络模块的基础类。 Sequential，序贯容器，类似Keras里面的序贯模型，另外这些是比较传统的卷积网络了。 Conv2d，二维卷积层。...

2018-05-17 00:59:49 8983

原创「Deep Learning」理解Pytorch中的「torch.utils.data」

Sina Weibo：小锋子Shawn Tencent E-mail：403568338@qq.com http://blog.csdn.net/dgyuanshaofeng/article/details/78761026基于Pytorch 0.4.0 四、DataLoader class torch.utils.data.DataLoader(dataset, batch_size...

2018-05-16 03:46:48 728

原创「Deep Learning」Note on ReLU

Sina Weibo：小锋子Shawn Tencent E-mail：403568338@qq.com http://blog.csdn.net/dgyuanshaofeng/article/details/80209602[1] Rectified Linear Units Improve Restricted Boltzmann Machines 2010 [paper]...

2018-05-12 02:20:00 123

原创「Deep Learning」Note on GELU（Gaussian Error Linear Unit）

Sina Weibo：小锋子Shawn Tencent E-mail：403568338@qq.com http://blog.csdn.net/dgyuanshaofeng/article/details/80209602[1] Bridging Nonlinearities and Stochastic Regularizers with Gaussian Error Linear...

2018-05-05 21:35:21 4503

原创「Deep Learning」Note on Softplus

Sina Weibo：小锋子Shawn Tencent E-mail：403568338@qq.com http://blog.csdn.net/dgyuanshaofeng/article/details/80209602[1] Rectified Linear Units Improve Restricted Boltzmann Machines 2010 [paper]...

2018-05-05 21:32:37 121

原创「Deep Learning」Note on SiLU（Sigmoid-weighted Linear Unit）

Sina Weibo：小锋子Shawn Tencent E-mail：403568338@qq.com http://blog.csdn.net/dgyuanshaofeng/article/details/80209602[1] Sigmoid-weighted Linear Units for Neural Network Function Approximation in Rei...

2018-05-05 21:28:20 1106

原创「Deep Learning」Note on Swish

Sina Weibo：小锋子Shawn Tencent E-mail：403568338@qq.com http://blog.csdn.net/dgyuanshaofeng/article/details/80208944这篇博客介绍比较新的激活函数——Swish，是ICLR2018 Workshop的文章。文章的结论就是“在任意神经网络，可以使用Swish单元替换ReLU单元。...

2018-05-05 20:07:20 294

原创「Deep Learning」Note on Machine Learning Yearning

Sina Weibo：小锋子Shawn Tencent E-mail：403568338@qq.com http://blog.csdn.net/dgyuanshaofeng/article/details/801983931 Why Machine Learning Strategy 为什么采用机器学习策略？因为机器学习是很多重要应用的基石。比如，web搜索，email反垃圾，...

2018-05-04 17:42:43 282

原创「Deep Learning」Note on SqueezeDet（挤压检测）

Sina Weibo：小锋子Shawn Tencent E-mail：403568338@qq.com http://blog.csdn.net/dgyuanshaofeng/article/details/80195971

2018-05-04 15:57:41 89

原创「Deep Learning」Note on MobileNet（移动网络）

Sina Weibo：小锋子Shawn Tencent E-mail：403568338@qq.com http://blog.csdn.net/dgyuanshaofeng/article/details/80195971

2018-05-04 15:48:39 130

原创「Deep Learning」Note on XNOR-Network（异或网络）

Sina Weibo：小锋子Shawn Tencent E-mail：403568338@qq.com http://blog.csdn.net/dgyuanshaofeng/article/details/80195971

2018-05-04 15:44:32 232

原创「Deep Learning」Note on SqueezeNet（挤压网络）

Sina Weibo：小锋子Shawn Tencent E-mail：403568338@qq.com http://blog.csdn.net/dgyuanshaofeng/article/details/80195971[1] SqueezeNet: AlexNet-level Accuracy with 50x Fewer Parameters and <0.5MB Mod...

2018-05-04 15:35:02 435

原创「Deep Learning」Note on WGAN

Sina Weibo：小锋子Shawn Tencent E-mail：403568338@qq.com http://blog.csdn.net/dgyuanshaofeng/article/details/80154161[1] Wasserstein Generative Adversarial Networks 2017 [paper]

2018-05-01 09:40:54 214

原创「Deep Learning」Note on GANs（生成对抗网络）

Sina Weibo：小锋子Shawn Tencent E-mail：403568338@qq.com http://blog.csdn.net/dgyuanshaofeng/article/details/80154103[1] Generative Adversarial Nets 2014 [paper]

2018-05-01 09:28:03 224

小锋子Shawn

原创「Deep Learning」Note on SFT-GAN（空间特征变换GAN）

原创「Deep Learning」Note on Super-FAN（超分人脸对齐网络）

原创「Machine Learning」Note on Neural Network and Random Forest（神经网络与随机森林）

原创「Deep Learning」Note on MaskLab

原创「Deep Learning」Note on Recurrent Salienct Transformation Network

原创「Deep Learning」Note on MSDNet（多尺度密集连接网络）

原创「Deep Learning」Note on NADM

原创「Deep Learning」Note on ADADELTA

原创「Deep Learning」Note on Adam

原创「Deep Learning」Note on RMSprop

原创「Deep Learning」Note on ADAGrad（比vanilla SGD好的优化算法）

原创「Deep Learning」Note on AMSGrad（比Adam好的优化算法）

原创「Pillow/PIL」Note on Image Module

原创「Deep Learning」理解Pytorch中的「torch.nn」

原创「Deep Learning」理解Pytorch中的「torch.utils.data」

原创「Deep Learning」Note on ReLU

原创「Deep Learning」Note on GELU（Gaussian Error Linear Unit）

原创「Deep Learning」Note on Softplus

原创「Deep Learning」Note on SiLU（Sigmoid-weighted Linear Unit）

原创「Deep Learning」Note on Swish

原创「Deep Learning」Note on Machine Learning Yearning

原创「Deep Learning」Note on SqueezeDet（挤压检测）

原创「Deep Learning」Note on MobileNet（移动网络）

原创「Deep Learning」Note on XNOR-Network（异或网络）

原创「Deep Learning」Note on SqueezeNet（挤压网络）

原创「Deep Learning」Note on WGAN

原创「Deep Learning」Note on GANs（生成对抗网络）

空空如也

空空如也