michaeljklein

├── LICENSE: 29 lines
├── README.md: 12 lines
├── Setup.hs: 2 lines
├── app
│   └── Main.hs: 5 lines
├── git-details.cabal: 82 lines
├── src
│   ├── Data
│   │   ├── Attoparsec
│   │   │   └── Text

michaeljklein

let configpath = '/Users/michaelklein/.cvimrc'
let homedirectory = '/Users/michaelklein'
set localconfig

let vimcommand = 'urxvt -e vim'
imap <C-o> editWithVim

set smoothscroll
" set noautofocus
set cncpcompletion

michaeljklein

#!/bin/sh

# To squash the last three commits: git squash 3
# It will open an interactive editor so you can 
#  more easily combine the comments

source "$(git --exec-path)/git-sh-setup"

USAGE="COMMITS"
function _squash() {

michaeljklein

unless Rails.env.production?
  require 'ftpd'
  require 'tmpdir'

  puts "****** loading tmp_ftp *********"

  class TmpFtp
    def initialize(temp_dir)
      @temp_dir = temp_dir
    end

michaeljklein

#!/bin/bash
# Create a post-checkout hook to show stashed changes 

cd .git/hooks&&("#!/bin/sh\nexec git stash list --pretty=oneline --abbrev-commit">post-checkout)&&chmod +x post-checkout

michaeljklein

class Mock
  attr_accessor :mock_original_name, :mock_call_stack

  def self.on(x)
    x.instance_eval do
      def method_missing(symbol, *args)
        puts "mocking #{self}.${symbol}(#{args.inspect})"
        @mock_call_stack ||= []
        @mock_call_stack << [symbol, args]

michaeljklein

Comparison:

from: markdown, to:   Fixnum, file:          test2.md, sum bytes              :   525983.2 i/s
from: markdown, to:   Fixnum, file:           test.md, sum bytes              :   127658.3 i/s -        4.12x  (±       0.14) slower
from:      org, to:   Fixnum, file:        format.org, sum bytes              :   123568.3 i/s -        4.26x  (±       0.14) slower
from: markdown, to:   Fixnum, file:   format.markdown, sum bytes              :   104980.1 i/s -        5.01x  (±       0.33) slower
from:      rst, to:   Fixnum, file:        format.rst, sum bytes              :    98854.8 i/s -        5.32x  (±       0.32) slower
from:  textile, to:   Fixnum, file:    format.textile, sum bytes              :    97966.4 i/s -        5.36x  (±       0.29) slower
from:    latex, to:   Fixnum, file:      format.latex, sum bytes              :    87885.2 i/s -        5.98x  (±       0.32) slower
from:    latex, to:   Fixnum, file:          bomb.tex, sum bytes              :    32267.8 i/s -       16.28x  (±     

michaeljklein

These benchmarks take about 45 minutes to run and perform a total of ~350k conversion (on my machine).

Here is the only significant win by pandoc-ruby:

Calculating -------------------------------------
from: latex, to: latex, file: bomb.tex, pandoc_rb
                          0.020  (± 0.0%) i/s -      1.000  in  49.770525s
from: latex, to: latex, file: bomb.tex, pandoc-ruby (read file)
                         21.586  (± 3.4%) i/s -    104.000  in   5.025571s

michaeljklein

https://developers.googleblog.com/2017/09/how-machine-learning-with-tensorflow.html

Even with an augmented and enriched training set, 
there is no substitute for images created by end-users in a variety of environmental conditions. 

Even with an augmented and enriched training set and test code, there is no substitute for code written and read by end-users in a varitety of conditions.

michaeljklein

 Obj <- _ -> Obj <- _ -> Nat <- _ -> Nat          Nat <- _ -> Nat <- _ -> Lit <- _ -> Lit
 ^           ^           ^           ^            ^           ^           ^           ^
 |           |           |           |            |           |           |           |
 _           _           _           _            _           _           _           _
 |           |           |           |            |           |           |           |
 v           v           v           v            v           v           v           v
 Obj <- _ -> Obj <- _ -> Nat <- _ -> Nat          Nat <- _ -> Nat <- _ -> Lit <- _ -> Lit
 ^           ^           ^           ^            ^           ^           ^           ^