aic25

#################################################################
# = This script transfers bash history to zsh history
# = Change bash and zsh history files, if you don't use defaults
#
# = Usage: ruby bash_to_zsh_history.rb
#
# = Author: Ankit Goyal
#################################################################

# change if you don't use default values

aic25

import json
import os

from tqdm import tqdm
from xmltodict import unparse

# BBOX_OFFSET: Switch between 0-based and 1-based bbox.
# The COCO dataset is in 0-based format, while the VOC dataset is 1-based.
# To keep 0-based, set it to 0. To convert to 1-based, set it to 1.
BBOX_OFFSET = 0

aic25

cmake_minimum_required( VERSION 3.6 )

# Create Project
project( test )
add_executable( freetype main.cpp )

# Set StartUp Project
set_property( DIRECTORY PROPERTY VS_STARTUP_PROJECT "freetype" )

# Set CharacterSet

aic25

#!/usr/bin/env python

# --------------------------------------------------------
# Faster R-CNN
# Copyright (c) 2015 Microsoft
# Licensed under The MIT License [see https://github.com/rbgirshick/py-faster-rcnn/blob/master/LICENSE for details]
# Written by Ross Girshick
# Modified by Sky Chon for webcam use
# --------------------------------------------------------

aic25

This gist contains a list of points I found very useful while watching the fast.ai "Practical deep learning for coders" and "Cutting edge deep learning for coders" MOOC by Jeremy Howard and team. This list may not be complete as I watched the video at 1.5x speed on marathon but I did write down as many things I found to be very useful to get a model working. A fair warning the points are in no particular order, you may find the topics are all jumbled up.

Before beginning, I want to thank Jeremy Howard, Rachel Thomas, and the entire fast.ai team in making this awesome practically oriented MOOC.

1. Progressive image resolution training: Train the network on lower res first and then increase the resolution to get better performance. This can be thought of as transfer learning from the same dataset but at a different resolution. There is one paper by NVIDIA as well that used such an approach to train GANs.

2. Cyclical learning rates: Gradually increasing the learning rate initially helps to avoid getting stuc

aic25

// Simple example of client.
// Client prints received messages to stdout and sends from stdin.

#include <errno.h>
#include <fcntl.h>
#include <stdio.h>
#include <signal.h>
#include <unistd.h>
#include <sys/select.h>
#include <netinet/in.h>

aic25

#
# Ubuntu Node.js Dockerfile
#
# https://github.com/dockerfile/ubuntu/blob/master/Dockerfile
# https://docs.docker.com/examples/nodejs_web_app/
#

# Pull base image.
FROM ubuntu:14.04