科学网

 找回密码
  注册
Instal Pytorch and related apps on Windows
高琳琳 2019-7-1 21:16
(一)Instal Pytorch Instal Anaconda Instal visual studio Download Cuda (local version) and instal. If 系统检测“ fails, then try to instal again 4. download cuDNN, and copy the following files: &nbs ...
个人分类: DeepLearning|1508 次阅读|没有评论
Pytorch training with multi GPUs
高琳琳 2019-5-27 22:00
When using DataParallel , with a list of gpu like , all the inputs that you give to the module have to be on dev0 . We can use the following codes to set with torch.cuda.device(dev0): t = t.cuda() From https://discuss. ...
个人分类: DeepLearning|1459 次阅读|没有评论
3D ResNet pre-trained on Kinetics
高琳琳 2019-5-27 21:00
注意:resnet-18与-34是shourtcuttype是A,其它models的shortcuttype是B resnet-18-kinetics.pth:--modelresnet--model_depth18--resnet_shortcutA resnet-34-kinetics.pth:--modelresnet--model_depth34--resnet_shortcutA ...
个人分类: DeepLearning|5104 次阅读|没有评论
Install Pytorch on Ubuntu
高琳琳 2019-5-26 17:29
instal Anaconda PS: if anacondda python is not the default python, you need to try the following methods to set the anaconda python as the default python 方法一:此方法需要用户具有sudo权限 (1) 在终端输入$sudo gedit /etc/profile,打开profile文件。 (2) 在文件末尾添加一行:export PATH=/ ...
个人分类: DeepLearning|1349 次阅读|没有评论
[转载]adaptive pooling
高琳琳 2019-5-26 13:45
torch.nn. AdaptiveAvgPool3d ( output_size ) Applies a 3D adaptive average pooling over an input signal composed of several input planes. The output is of size D x H x W, for any input size. The number of output features is equal to the number of input planes. https://pyto ...
个人分类: DeepLearning|3682 次阅读|没有评论
[转载]DL tricks
高琳琳 2019-5-17 21:42
normalization using pre-trained vgg Q: If I wanna extract features with VGG (pre-trained on the ImageNet) on my dataset, should I subtract the ImageNet mean, or should I calculate my dataset's mean and std firstly? Are there big differences between the ImageNet and other RGB datasets general ...
个人分类: DeepLearning|1209 次阅读|没有评论
Atom如何使用go to definition
高琳琳 2019-5-16 12:23
安装三个包:symbol-gen, symbols-view与goto-definition 重启atom https://blog.csdn.net/wowotuo/article/details/78447704
个人分类: DeepLearning|2674 次阅读|没有评论
[转载]Anaconda installation and path configuration on windows
高琳琳 2019-4-29 21:41
Download Anaconda from the official website Select the default options Check whether Jupyter can be run successfully by entering jupyter notebook 4. If you didn’t check the add Anaconda to path argument during the installation process, you will have to add python and conda t ...
个人分类: DeepLearning|1796 次阅读|没有评论
[转载]Residual blocks--Building blocks of ResNet
高琳琳 2019-4-28 21:15
Understanding a residual block is quite easy. In traditional neural networks, each layer feeds into the next layer. In a network with residual blocks, each layer feeds into the next layer and directly into the layers about 2–3 hops away. That’s it. But understanding the intuition behind why it w ...
个人分类: DeepLearning|2370 次阅读|没有评论
[转载]基于弱监督的视频时序动作检测的介绍
高琳琳 2019-2-9 16:46
From:机器学习研究会订阅号 关于untrimmed video analysis(未剪辑视频分析)的领域,在众多大牛的努力下(@林天威、@Showthem、@高继扬( 高博曾在极市做过关于时序动作检测的线上分享,点击直接跳转阅读 )等),前两年基于全监督的视频时序检测一路高歌猛进,直至今年的 TAL-Net (h ...
个人分类: DeepLearning|4863 次阅读|没有评论

Archiver|手机版|科学网 ( 京ICP备07017567号-12 )

GMT+8, 2024-4-25 00:30

Powered by ScienceNet.cn

Copyright © 2007- 中国科学报社

返回顶部