Skip to content

Instantly share code, notes, and snippets.

View Logrus's full-sized avatar

Logrus Logrus

19 followers · 8 following
View GitHub Profile
@Logrus
Logrus / CMakeLists.txt
Created April 5, 2014 07:27
PCL Viewer with Qt GUI minimal code
cmake_minimum_required(VERSION 2.6)
PROJECT(qt_vtk_pcl)
list(APPEND CMAKE_MODULE_PATH ${CMAKE_CURRENT_SOURCE_DIR})
find_package(PCL 1.7 REQUIRED)
include_directories(${PCL_INCLUDE_DIRS})
link_directories(${PCL_LIBRARY_DIRS})
@Logrus
Logrus / sampleAndReturnlist.cpp
Last active August 29, 2015 14:17
struct point{
unsigned int x, y;
point() {};
point (int a, int b) {
x = a; y = b;
}
bool point::operator ==(const point &a){
return (x == a.x && y == a.y);
}
};
@Logrus
Logrus / .vimrc
Last active October 19, 2016 09:46
vimconf
" Installing vundle, run:
" git clone https://github.com/VundleVim/Vundle.vim.git ~/.vim/bundle/Vundle.vim
" after, run in vim
" :PluginInstall
"
set nocompatible " be iMproved, required
filetype off " required
set rtp+=~/.vim/bundle/Vundle.vim
call vundle#begin()
@Logrus
Logrus / .tmux.conf
Last active November 24, 2020 09:08
# Add to .bashrc
# alias tmux="TERM=screen-256color-bce tmux"
# Installing new tmux on 14.04
# sudo add-apt-repository ppa:pi-rho/dev
# sudo apt-get update
# sudo apt-get install python-software-properties software-properties-common
# sudo apt-get install -yqq tmux-next=2.3~20161028~bzr3606+20-1ubuntu1~ppa0~ubuntu14.04.1
# Add to .bashrc
# alias tmux="TERM=screen-256color-bce tmux-next"
@Logrus
Logrus / test_endianness.cc
Created October 30, 2016 18:27
// Source: http://stackoverflow.com/questions/12791864/c-program-to-check-little-vs-big-endian
#include <stdio.h>
int main()
{
int x = 1;
char *y = (char*)&x;
if(*y){
printf("This computer is LITTLE endian.\n");
printf(" higher memory \n");
@Logrus
Logrus / gist:16412952fc890f16f289de3d8e08fd2c
Created November 21, 2016 18:22
plot_loss_py.sh
#!/bin/bash
cat *log | grep -o ".Loss.[0-9]*\.[0-9]*" | grep -o "[0-9]*\.[0-9]*" > loss_values.txt
num=51
if [ $# -eq 1 ]; then
num=$1
fi
python -c "import numpy as np; from scipy.signal import savgol_filter; y=np.loadtxt('loss_values.txt'); yh=savgol_filter(y, $num, 3); np.savetxt('smoothed_values.txt',yh);"
(
gnuplot <<- EOF
@Logrus
Logrus / gist:7e20e9e2e795f5da06c4952d3fc2524b
Created November 21, 2016 18:23
plot_loss.sh
cat *.log | grep -o ".Loss.[0-9]*\.[0-9]*" | grep -o "[0-9]*\.[0-9]*" > loss_values.txt
gnuplot <<- EOF
set terminal dumb
set xlabel "Iterations"
set ylabel "Loss"
plot "loss_values.txt"
EOF
@Logrus
Logrus / bash_helpers.sh
Created November 15, 2019 23:58
showasm() {
$1++ $2 -O$3 -std=c++2a -S -masm=intel -o _temp_name.s && cat _temp_name.s | c++filt | grep -vE '\s+\.' > _temp_name_.s && rm _temp_name.s && vim _temp_name_.s
}
@Logrus
Logrus / extract_images_from_bagfile.py
Created July 20, 2020 16:05
Dump sensor_msgs/CompressedImage from bagfile into a folder as png or jpeg files
#!/usr/bin/env python3
"""Extract sensor_msgs/CompressedImage from bagfile and save into a folder as png or jpeg files
Example usage:
./extract_images_from_bagfile.py bagfile.bag "/camera_topic" output_folder/images
Then files will be saved inside:
output_folder/images/image_1594120767.5438294.jpeg
output_folder/images/image_1594120767.8838294.jpeg
output_folder/images/image_1594120767.9538294.jpeg
@Logrus
Logrus / image_learning.py
Last active September 13, 2023 13:36
Memorize an image with NN, made to play around with sin activation, positional encodings and etc.
"""
Recently Implicit Neural Representations gain popularity.
One of the issues there though is that the learned representations
are low-frequency biased, resulting in over-smoothed representations.
There has been a few approaches suggested to alleviate the issue,
for example by using positional encodings.
An alternative could be using Sin/Cos activation functions,
which in essence present a learnable basis functions.