wangkuiyi

# This script download, build and deploy MPICH 3.1
# (http://www.mpich.org) and pLDA 3.1 (https://code.google.com/p/plda)
# to computers in a cluster.
#
# Usage:
#
# You download and run this script from any directory.  For example,
# consider that Alice is running this script to install MPICH and pLDA
# on David's workstation in the role of Steven, she should do:
#

wangkuiyi

curl -u _token:$(gcloud auth print-access-token)   https://gcr.io/v2/tensorflow/tensorflow/tags/list
docker run --rm -it gcr.io/tensorflow/tensorflow:r0.9rc0 /bin/bash

wangkuiyi

digraph CI {
  github [label="github.com"];
  ngrok1 [label="ngrok tunnel 1"];
  ngrok2 [label="ngrok tunnel 2"];
  config1 [label="configuration 1"];
  config2 [label="configuration 2"];
  
  github -> ngrok1 -> config1;
  github -> ngrok2 -> config2;
}

wangkuiyi

cc_plugin(
    name = 'lda',
    srcs = ['lda.c'],
    deps = ['#pthread']
)

wangkuiyi

project="paddlepaddle/paddle"
if [[ $# -ne 0 ]]; then
   project=$1
fi

file=${project/\//-}
> $file.issues

issues_per_page=25
page=1

wangkuiyi

project="wangkuiyi/mapreduce-lite"
if [[ $# -ne 0 ]]; then
   project=$1
fi

issues_per_page=25
page=1
until [ $issues_per_page -lt 25 ]; do
    echo Parsing page: $page >&2

wangkuiyi

PaddlePaddle is Launching New API

PaddlePaddle is a deep learning framework originally developed in Baidu four years ago. Experienced Baidu researchers and engineers have successfully applied it in 80% of Baidu products. When we open sourced PaddlePaddle in September 2016, we know it is time to make it easy-to-use by everyone who is interested in deep learning.

Today we are happy to announce the alpha release of our new API. And we updated example programs in our open source book Deep Learning with PaddlePaddle accordingly.

The following screenshot shows that a PaddlePaddle program using the new API could be much shorter than its old counterpart.

wangkuiyi

for i in $(du -a | cut -f 2); do
    if [[ -d $i ]]; then
	echo === $i |
	    sed 's/\.//';
	grep '#include <majel'  $i/*.{h,cu,cc,cpp,hpp,hh} 2>/dev/null |
	    sed 's/^.*#include <majel//' |
	    sed 's/\/[-_a-z0-9]*.h>$//' |
	    sort |
	    uniq;
    fi;

wangkuiyi

#include <iostream>

class Image {
};
  
class blur {
 public:
  blur(double factor) : factor_(factor) {}
  
  void operator()(Image & img) const   {

wangkuiyi

#include <iostream>

namespace strings {
void Print() {
  std::cout << "strings::Print" << std::endl;
}
}  // namespace strings

namespace uses {
namespace detail {