- 博客(8)
- 资源 (2)
- 收藏
- 关注
原创 torch 重装 升级
I suggest that you cleanup any installation of torch first, using this script:https://github.com/torch/ezinstall/raw/master/clean-old.sh要删除旧文件。
2015-06-30 19:00:19 1826
原创 xavierFiller 体会
之前用caffe时,权值是用的xavierFiller填充的,配合以比较小的学习速率(SGD),发现效果出奇好,随便配置下网络,就能干到94%+的准确率,不论训练集还是测试集。不知道原因找的准不准,贴出配置,供查阅。下面是solver的配置:# The train/test net protocol buffer definitionnet: "./car_detector_tr
2015-06-21 14:58:18 3037 1
原创 torch7 nn 模块 ClassNLLCriterion
nn模块中的loss函数有一个选择就是:nn.ClassNLLCriterion。既The negative log likelihood criterion。需要说明的是,其表达式criterion = nn.ClassNLLCriterion([weights]) 中,weights是用于类别加权的,为一个1维的Tensor。============模块自带文档============
2015-06-16 10:46:28 3873
原创 Torch7 15分钟学习文档 记录
http://tylerneylon.com/a/learn-lua/ 一段带有丰富注释的代码。Only nil and false are falsy; 0 and '' are true!调试环境根据建议,用eclips。感觉代码风格有点像matlab和python的混血,看来前段时间的python没白啃。
2015-06-11 09:37:03 1408 1
原创 caffe, torch, theano
http://fastml.com/torch-vs-theano/caffe for application.torch and theano for research.since i already spent a lot of time in caffe, I prefer go on with caffe until cant.
2015-06-10 10:38:04 974 1
原创 caffe prototxt 编写注意
https://github.com/BVLC/caffe/blob/dev/src/caffe/proto/caffe.proto这个是要参考的,里面有各种参数的说明。layer { name: "car_detector" type: "Data" top: "data" top: "label" include { phase: TRAIN
2015-06-09 13:24:53 4827
原创 google.protobuf.... modul not found
caffe is using anaconda, so, make sure the PATH you are using.cant find google.protobuf means you should install it with pip, the pip provided by the anaconda.if "pip install protobuf" failed, wit
2015-06-08 13:20:27 2178
修复移动硬盘文件目录损坏且无法读取ver2.0 – Chaos Club.pdf
2020-02-22
Real-time hand gesture recognition from depth images
2014-06-08
空空如也
TA创建的收藏夹 TA关注的收藏夹
TA关注的人