abunsen

./minerd -a cryptonight -o stratum+tcp://etnpool.minekitten.io:3333 -u addy -p x -t 4

abunsen

Hello Andrew Trask!

I'm reading "Grokking deep learning" - I've enjoyed chapters 1-4 but Chapter 5's speed, mistakes and omitted materials have lead to a bit of a roadblock for me.

I believe I've managed to create a simple neural net, that uses gradient descent to learn but when I use the raw MNIST dataset, without transforming the pixel values from large numbers (numbers above 1) to binary (0,1) then I can't train my model, I get "double scalar" errors and all sorts of weird things (see below for examples).

I was wondering if you could look over my code, tell me where I've went wrong (tweet me: @bunsen or email: auston.bunsen@gmail.com) & if you'd like you can use this code for free, under the MIT license.

abunsen

require "httparty"
require 'nokogiri'

homepage_url = "https://miami.craigslist.org/"

response1 = HTTParty.get homepage_url
home_page = Nokogiri::HTML(response1.body)
cats = []
url_tag = []
jobs = []

abunsen

.slide  {
  background-image: url(https://wyncode-report.surge.sh/images/background-eee688ff801cbdd8e7b64eb43a99b9fa.jpg);
}

.slide h1 {
  color: white !important;
  text-align: center;
}

abunsen

# Configure colors, if available.
if [ -x /usr/bin/tput ] && tput setaf 1 >&/dev/null; then
    c_reset='\[\e[0m\]'
    c_user='\[\e[0;32m\]'
    c_path='\[\e[1;34m\]'
    c_git_clean='\[\e[0;37m\]'
    c_git_staged='\[\e[0;32m\]'
    c_git_unstaged='\[\e[0;31m\]'
else
    c_reset=

abunsen

"""
Minimal character-level Vanilla RNN model. Written by Andrej Karpathy (@karpathy)
BSD License
"""
import numpy as np

# data I/O
data = open('input.txt', 'r').read() # should be simple plain text file
chars = list(set(data))
data_size, vocab_size = len(data), len(chars)

abunsen

input_data = ['1', '4 5', 'S V 1', 'S W 4', 'V W 2', 'V T 6', 'W T 3', 'S']
my_graph = {
	'A': {'B': 7, 'C': 8},
	'B': {'A': 7, 'F': 2},
	'C': {'A': 8, 'F': 6, 'G': 4},
	'D': {'F': 8},
	'E': {'H': 1},
	'F': {'B': 2, 'C': 6, 'D': 8, 'G': 9, 'H': 3},
	'G': {'C': 4, 'F': 9},
	'H': {'E': 1, 'F': 3}

abunsen

So I built an algorithm that tries to lose tictactoe no matter what. If you'd like to battle your algorithm against mine, I think it would be fun. This is the spec of what your server should accept & return. If you built an algorithm that can beat it, please tweet me @bunsen.

Your server should accept GET requests with the following parameters:

• row
• col
• mark
• grid - will be a url encoded array, like this: [["x",%20"x",%20nil],%20[nil,%20nil,%20"o"],%20[nil,%20"o",%20nil]]

It should return JSON that looks like this:

abunsen

Xvfb :99 -ac -screen 0 1024x768x8

abunsen

I’m writing this from the AT&T Mobile App Hackathon, which really is the highlight of this week. It has turned out really well - over 200 attendees, lots of interesting ideas pitched, and bunch of impressive apps hacked together very quickly. 17 teams total made it to the final presentation. This is the largest hackathon I’ve seen in Miami so far, with the Startup Weekend put together by Marc Billings of Digiport/Incubate Miami probably the second largest at ~100 or so attendees and 10 teams. The Starting Gate has a nice writeup of the first and final evenings of the event, Kristofer Rios covered it in the Herald as well, and Startropica has a detaled account of all 3 days.

Earlier in the week, there was an interesting exchange of opinion on the state and future of the Miami tech ecosystem. Juan Pablo Cappello, a latam serial entrepreneur, investor and attorney, posted an opinion titled “Let’s restart our engines and accelerate Miami’s tech-hub potential”

Auston Bunsen, the geekpreneur behind SuperConf (*