Skip to content

Instantly share code, notes, and snippets.

View rahimnathwani's full-sized avatar

Rahim Nathwani rahimnathwani

41 followers · 106 following
View GitHub Profile
@rahimnathwani
rahimnathwani / trim.py
Created March 24, 2012 03:33
Trim the second field in a CSV file down to 3999 characters
import csv
csvReader = csv.reader(open('input.csv','rb'), delimiter=',',quotechar='"')
csvWriter = csv.writer(open('output.csv','wb'), delimiter=',',quotechar='"',quoting=csv.QUOTE_MINIMAL)
for row in csvReader:
csvWriter.writerow([row[0],row[1][0:3998],row[2],row[3],row[4],row[5],row[6],row[7],row[8],row[9],row[10],row[11],row[12]])
@rahimnathwani
rahimnathwani / gist:5107454
Created March 7, 2013 11:37

Twitter公式クライアントのコンシューマキー

Twitter for iPhone

Consumer key: IQKbtAYlXLripLGPWd0HUA
Consumer secret: GgDYlkSvaPxGxC4X8liwpUoqKwwr3lCADbz8A7ADU

Twitter for Android

Consumer key: 3nVuSoBZnx6U4vzUxf5w
Consumer secret: Bcs59EFbbsdF6Sl9Ng71smgStWEGwXXKSjYvPVt7qys

Twitter for iPad

Consumer key: CjulERsDeqhhjSme66ECg

@rahimnathwani
rahimnathwani / gist:861fbac9b22d64242369
Created July 17, 2014 14:57
Split audio file on silence
sox -V3 output.wav out.wav silence -l 0 1 0.2 2% : newfile : restart
@rahimnathwani
rahimnathwani / frequency_estimator.py
Created July 20, 2014 11:46 — forked from endolith/frequency_estimator.py
from __future__ import division
from scikits.audiolab import flacread
from numpy.fft import rfft, irfft
from numpy import argmax, sqrt, mean, diff, log
from matplotlib.mlab import find
from scipy.signal import blackmanharris, fftconvolve
from time import time
import sys
from parabolic import parabolic
@rahimnathwani
rahimnathwani / gist:57f1a5743566d85e9a4c
Last active October 12, 2015 04:31 — forked from andrewlkho/gist:23b5b5bcf853b3473421
How to install debian to a USB drive for use on a MacBook Air

This guide shows how to install debian so that it can be booted and run from a USB drive on a MacBook Air. I use this to store a copy of my PGP master keypair. Networking is deliberately not configured and all operations requiring the master keypair (such as signing other people's keys) are done only on this system. See the debian wiki for information on how to separate your key. Note that this is slightly different to using a live CD. It is not possible to tailor the base configuration of a live CD to one's own requirements, and they frequently automatically connect to the nearest local network. Nevertheless, if you wish to use a live CD instead, I recommend Tails. Arturo Filastò has written a similar guide for Tails.

The method I use requires [VirtualBox][4], which should also contain a working debian installation itself. An alternative method would be to use [debootstrap][5] b

@rahimnathwani
rahimnathwani / gce-create.sh
Created March 20, 2017 12:09
Create GPU instance on Google Cloud, for course.fast.ai
gcloud beta compute instances create gpu-instance-1 \
--machine-type n1-standard-1 --zone asia-east1-a \
--accelerator type=nvidia-tesla-k80,count=1 \
--image-family ubuntu-1604-lts --image-project ubuntu-os-cloud \
--maintenance-policy TERMINATE --no-restart-on-failure \
--boot-disk-size=25GB
@rahimnathwani
rahimnathwani / gce-create-from-snapshot.sh
Created March 20, 2017 12:11
Create GPU instance from previously-create snapshot
gcloud compute snapshots list
gcloud compute disks create deep-learning-persistent-disk-1 --size=25GB --source-snapshot=deep-learning-snapshot-1
gcloud beta compute instances create gpu-instance-1 \
--machine-type n1-standard-1 --zone asia-east1-a \
--accelerator type=nvidia-tesla-k80,count=1 \
--image-family ubuntu-1604-lts --image-project ubuntu-os-cloud \
--maintenance-policy TERMINATE --no-restart-on-failure \
--disk=auto-delete=yes,boot=yes,name=deep-learning-persistent-disk-1
gcloud compute instances add-tags gpu-instance-1 --tags https-server
@rahimnathwani
rahimnathwani / install-gpu-gce.sh
Last active August 4, 2017 19:28
Set up a Google Cloud GPU instance for doing the exercises in course.fast.ai
#!/bin/bash
if ! dpkg-query -W cuda; then
curl -O http://developer.download.nvidia.com/compute/cuda/repos/ubuntu1604/x86_64/cuda-repo-ubuntu1604_8.0.61-1_amd64.deb
dpkg -i ./cuda-repo-ubuntu1604_8.0.61-1_amd64.deb
fi
apt-get update
apt-get -y upgrade
apt-get -y install tmux build-essential gcc g++ make binutils software-properties-common cuda unzip
modprobe nvidia
nvidia-smi
@rahimnathwani
rahimnathwani / rtsp_server_build.sh
Created November 16, 2017 13:15
cd /usr/src/sn986
./sdk.unpack
cd /usr/src/sn986/snx_sdk/buildscript
make sn98660_QR_Scan_402mhz_sf_defconfig
cd /usr/src/sn986/snx_sdk/middleware
mkdir -p _install/lib _install/include;
cp video/middleware/lib/lib* rate_ctl/middleware/lib/lib* sdrecord/middleware/lib/lib* audio/middleware/lib/lib* _install/lib
cp -r video/middleware/include/snx_isp video/middleware/include/snx_vc rate_ctl/middleware/include/snx_rc sdrecord/middleware/include/snx_record audio/middleware/include/* _install/include
PATH=/usr/src/sn986/snx_sdk/toolchain/crosstool-4.5.2/bin:${PATH}
cd /usr/src/sn986/snx_sdk/app/example/src/ipc_func/rtsp_server
@rahimnathwani
rahimnathwani / gist:b3677d8c6d9a908257fb846e0b78c7cb
Last active October 10, 2018 16:45 — forked from zjx20/gist:1c7f4dd1392866f9d8b00dd2a05bd481
kcptun install and systemd daemon
# wget https://github.com/xtaci/kcptun/releases/download/v20170525/kcptun-linux-amd64-20170525.tar.gz
# tar xvf kcptun-linux-amd64-20170525.tar.gz
mv server_linux_amd64 /usr/local/bin/kcptun_server
mv /usr/local/bin/server_linux_amd64 /usr/local/bin/kcptun_server
sudo mkdir -p /etc/kcptun
sudo bash -c "cat <<EOT > /etc/kcptun/server_conf.json
{
\"listen\": \":21\",
\"target\": \"127.0.0.1:20001\",