woudsma

// Simulates try/catch statements wrapping async functions in (nested) object
// Returns copy of original object with basic promise-rejection handling for async functions using .catch
// Saves the time of writing try/catch statements in every single async function

const trycatch = obj => Object.keys(obj)
  .reduce((acc, curr) => ({ ...acc, [curr]: obj[curr] instanceof Function
    ? (...args) => obj[curr](...args).catch(::console.error)
    : trycatch(obj[curr]) }), {})

woudsma

Kirby + AutoGit on Dokku

TL;DR

• Configure Kirby for Dokku
• Install deployment-keys and host keys Dokku plugins
• Set up staging and production environments
• Clone project repository into apps persistent storage folders
• Mount desired folders including .git folder to apps
• Add GIT_DIR and GIT_WORK_TREE environment variables to containers
• Deploy to Dokku

woudsma

Install Homebrew

• xcode-select --install
• /usr/bin/ruby -e "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/master/install)"

Install PostgreSQL + PostGIS

• brew update
• brew install postgres
• brew install postgis
• brew services start postgres
• createdb paperstorm
• psql

woudsma

TLS secured TCP exposed Docker daemon on Dokku host - setup

1. Create certificates
2. Edit Docker options
3. Restart Docker
4. Copy client certificates from host
5. (optional) Add remote endpoint in Portainer

Tested on a standard $5/mo DigitalOcean VPS running Ubuntu 16.04.

---

woudsma

Using the PostGIS extension on a PostgreSQL database with Dokku

docker pull mdillon/postgis:latest  
export POSTGRES_IMAGE="mdillon/postgis"  
export POSTGRES_IMAGE_VERSION="latest"  
dokku postgres:create my-postgis-db

woudsma

npm i k-d-tree

const kd = require('k-d-tree')

const data = new Array(1000)
  .fill()
  .map(() => ({
 type: "Point",

woudsma

#!/bin/bash

echo "Compositing files and ramping up video framerate"
echo "================================================"

rm -rf render && mkdir render

for file in *.mov; do
	# Ramp up video framerate, add audio
	ffmpeg \

woudsma

Retrain a MobileNet model and use it in the browser with TensorFlow.js

DRAFT

Combining TensorFlow for Poets and TensorFlow.js.
Retrain a MobileNet V1 or V2 model on your own dataset using the CPU only.
I'm using a MacBook Pro without Nvidia GPU.

MobileNets can be used for image classification. This guide shows the steps I took to retrain a MobileNet on a custom dataset, and how to convert and use the retrained model in the browser using TensorFlow.js. The total time to set up, retrain the model and use it in the browser can take less than 30 minutes (depending on the size of your dataset).

Example app - HTML/JS and a retrained MobileNet V1/V2 model.

woudsma

import math
from bigfloat import *

class LazyCartesianProduct:
    def __init__(self, sets, context):
        self.sets = sets
        self.context = context
        self.divs = []
        self.mods = []
        self.maxSize = BigFloat.exact(1)

woudsma

Dokku ACL plugin

# Dokku host
dokku plugin:install https://github.com/dokku-community/dokku-acl.git acl

dokku ssh-keys:add guest guest-key.pub
dokku ssh-keys:list

mkdir ~dokku/.dokkurc
echo "export DOKKU_ACL_ALLOW_COMMAND_LINE=1" >> ~dokku/.dokkurc/acl