Qingqing Cao csarron

Scrcpy Shortcuts

Action	Shortcut
switch fullscreen mode	`Ctrl`+`f`
resize window to 1:1 (pixel-perfect)	`Ctrl`+`g`
resize window to remove black borders	`Ctrl`+`x` \| Double-click¹
click on `HOME`	`Ctrl`+`h` \| Middle-click
click on `BACK`	`Ctrl`+`b` \| Right-click²
click on `APP_SWITCH`	`Ctrl`+`m`

MobileNet Inference Time

Movidius NCS Accelerator

Using NCSDK 1.12.00, lastest as of April 2018

Version	Average Inference Time

Rank (duplicates merged)	Model	Technique	Paper	Notes
1	MARS	BiLSTM + attn	https://posts.careerengine.us/p/5ad8637755ba6b136fc793ad	(slides in English, notes in Chinese)
2	QANet	attn	https://arxiv.org/abs/1804.09541
3	Reinforced Mnemonic Reader	BiLSTM	https://arxiv.org/abs/1705.02798
4	r-net	BiGRU	https://www.microsoft.com/en-us/research/wp-content/uploads/2017/05/r-net.pdf
5	SLQA+	BiLSTM	http://www.aclweb.org/anthology/P18-1158
6	BiDAF + Self Attention + ELMo + A2D	BiGRU	https://arxiv.org/abs/1710.10723
7	Hybrid AoA Reader	BiGRU	https://arxiv.org/abs/1607.04423
8	MAMCN+	BiGRU	http://www.aclweb.org/anthology/W18-2603
9	MDReader	unknown

install prebuilt eggs:

Ubuntu: may need to export libjvm.so ls -l /etc/alternatives/java will output something like: /etc/alternatives/java -> /usr/lib/jvm/java-8-openjdk-amd64/jre/bin/java then do: export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/lib/jvm/java-8-openjdk-amd64/jre/lib/amd64/server/ export JCC_JDK=/usr/lib/jvm/java-8-openjdk-amd64

for macOS: easy_install JCC-3.2-py3.6-macosx-10.13-x86_64.egg lucene-7.5.0-py3.6-macosx-10.13-x86_64.egg

	if (!([Security.Principal.WindowsPrincipal][Security.Principal.WindowsIdentity]::GetCurrent()).IsInRole([Security.Principal.WindowsBuiltInRole] "Administrator")) { Start-Process powershell.exe "-NoProfile -ExecutionPolicy Bypass -File `"$PSCommandPath`"" -Verb RunAs; exit }

	if (!(Get-Command choco -errorAction SilentlyContinue))
	{
	Set-ExecutionPolicy -Scope CurrentUser -ExecutionPolicy RemoteSigned -Confirm:$false
	"choco not exists or not installed, begin installing chocolatey..."
	$script = New-Object Net.WebClient
	$script \| Get-Member
	$script.DownloadString("https://chocolatey.org/install.ps1")
	iwr https://chocolatey.org/install.ps1 -UseBasicParsing \| iex

	# code $profile.CurrentUserAllHosts
	Import-Module posh-git
	function ga {
	git add $args
	}
	function gaa {
	git add --all
	}
	function gb {
	git branch

	use feed dict instead of directly pass tensor
	embeddings = self.session.run(self.encoding_tensor, feed_dict={
	self.input_placeholder: need_compute
	})

	# si_emb = self.session.run(self.embed([s]))[0]

	RCol='\e[0m' # Text Reset

	# Regular Bold Underline High Intensity BoldHigh Intens Background High Intensity Backgrounds
	Bla='\e[0;30m'; BBla='\e[1;30m'; UBla='\e[4;30m'; IBla='\e[0;90m'; BIBla='\e[1;90m'; On_Bla='\e[40m'; On_IBla='\e[0;100m';
	Red='\e[0;31m'; BRed='\e[1;31m'; URed='\e[4;31m'; IRed='\e[0;91m'; BIRed='\e[1;91m'; On_Red='\e[41m'; On_IRed='\e[0;101m';
	Gre='\e[0;32m'; BGre='\e[1;32m'; UGre='\e[4;32m'; IGre='\e[0;92m'; BIGre='\e[1;92m'; On_Gre='\e[42m'; On_IGre='\e[0;102m';
	Yel='\e[0;33m'; BYel='\e[1;33m'; UYel='\e[4;33m'; IYel='\e[0;93m'; BIYel='\e[1;93m'; On_Yel='\e[43m'; On_IYel='\e[0;103m';
	Blu='\e[0;34m'; BBlu='\e[1;34m'; UBlu='\e[4;34m'; IBlu='\e[0;94m'; BIBlu='\e[1;94m'; On_Blu='\e[44m'; On_IBlu='\e[0;104m';
	Pur='\e[0;35m'; BPur='\e[1;35m'; UPur='\e[4;35m'; IPur='\e[0;95m'; BIPur='\e[1;95m'; On_Pur='\e[45m'; On_IPur='\e[0;105m';
	C

	import itertools

	def grouper(iterable, n, fillvalue=None):
	"""Collect data into fixed-length chunks or blocks"""
	# grouper('ABCDEFG', 3, 'x') --> ABC DEF Gxx
	args = [iter(iterable)] * n
	return itertools.zip_longest(*args, fillvalue=fillvalue)

	a=itertools.chain.from_iterable(itertools.repeat(['a', 'b', 'c'], 4))
	b=itertools.chain.from_iterable(itertools.repeat(['d', 'e', 'f', 'g'], 3))

	''' Script for downloading all GLUE data.

	Note: for legal reasons, we are unable to host MRPC.
	You can either use the version hosted by the SentEval team, which is already tokenized,
	or you can download the original data from (https://download.microsoft.com/download/D/4/6/D46FF87A-F6B9-4252-AA8B-3604ED519838/MSRParaphraseCorpus.msi) and extract the data from it manually.
	For Windows users, you can run the .msi file. For Mac and Linux users, consider an external library such as 'cabextract' (see below for an example).
	You should then rename and place specific files in a folder (see below for an example).

	mkdir MRPC
	cabextract MSRParaphraseCorpus.msi -d MRPC