里長伯活動佈告欄

最近在研究Intel Caffe  (前文提要： 1.  Machine Learning 筆記整理 名詞篇   2.  How to install Intel Caffe on CentOS 筆記  ) 於是就開始了CentOS 的摸索，然後在CentOS 上安裝 OpenCV的時候遇到 cmake 版本太舊的問題 ?! 於是就來寫這篇 如何在CentOS 手動更新 cmake 1.  下載 可以到這個網址，找版本 https://cmake.org/files/ 我想升到 2.8.12 就到 https://cmake.org/files/v2.8/ 這邊來找 2. 安裝 # tar -zxvf cmake-2.8.12 # cd cmake-2.8.12 # ./bootstrap # make # make install 就對了 !!

最近在研究Intel Caffe 所以筆記人生就開始了新的篇章； We will talk about "How to install Intel Caffe on CentOS" 1. Down load Intel Caffe from github  https://github.com/intelcaffe sudo yum install git git clone git@github.com:intelcaffe/caffe.git or  sudo yum install wget wget https://github.com/intelcaffe/caffe/archive/master.zip 2. General dependencies please visit RHEL / Fedora / CentOS Installation  http://caffe.berkeleyvision.org/install_yum.html sudo yum install protobuf-devel leveldb-devel snappy-devel opencv-devel boost-devel hdf5-devel and sudo yum install gflags-devel glog-devel lmdb-devel 這樣會有基本的安裝，但實際上有遇到很多問題，先烈出來再來好好整理。 像是安裝少套件 可以來這邊找 Download the latest epel-release rpm from http://dl.fedoraproject.org/pub/epel/7/x86_64/ Install epel-release rpm: # rpm -Uvh epel-release*rpm Install leveldb-devel rpm package:  # yum install leveldb-devel 然後要開始Make前請看這篇 Caffe can be compiled with either Make or CMake. Make is officially supported while CMake is supported by the community. http://caffe

看到Google I/O 上提出的Google Home 以及Amazon 的Echo，於是找了一下如何用Intel Edison 做語音辨識。 關鍵字除了 Voice Recognition 外，其他還有 STT (Speech to Test) 及 Speech Recognition 快速找到了 Instructables  Speech to Text 主要利用 Python Quickstart: pip install SpeechRecognition 詳見  SpeechRecognition Library for performing speech recognition, with support for several engines and APIs, online and offline. Speech recognition engine/API support: CMU Sphinx   (works offline) Google Speech Recognition Wit.ai Microsoft Bing Voice Recognition api.ai IBM Speech to Text 看來只有CMU 佛心來的提供local 端辨識。其他都是線上版本。 前文提到的Amazon echo  是用  Amazon Alexa service  又稱AVS (Alexa Voice Service) 在github上也可以找到用Intel Edison 接 Amazon Alexa Amazon Alexa for Intel Edison added camera hand gesture   在Youtube上也有些影片討論：

今天Google I/O 展示了Google Home 要跟Amazon Echo 對打。 趁機找了一下 Intel的語音辨識 發現以下資訊： Intel® RealSense™ Voice Recognition Sample Application Intel® RealSense™ Voice Technologies 裡面有提到 Dragon Assistant with Natural Language Understanding

CNN convolutional neural network (CNN, or ConvNet)  DNN deep neural network (DNN) MKL Intel® Math Kernel Library ( Intel® MKL ) DAAL  Intel® Data Analytics Acceleration Library ( Intel® DAAL ) Caffe   Caffe  is a deep learning framework developed by Berkeley Vision and Learning Center (BVLC) and one of the most popular community frameworks for image recognition. Together with  AlexNet , a neural network topology for image recognition, and  ImageNet , a database of labeled images, Caffe is often used as a benchmark. Intel® Modern Code Multi-level parallelism is a framework that uses all of the parallel performance features available on modern hardware via vectorization, multi-threading, and multi-node optimizations. Explore how to deliver multi-level parallel algorithms that effectively scale forward for today’s and tomorrow’s hardware. 相關文章： Accelerating Machine Learning and Deep Learning with Intel Tools and Libraries Single Node Caffe Scoring and Training on Intel® Xeon E5-Series Pro