dmichael

# Mocha.js test runner support
# https://github.com/jfirebaugh/konacha

Konacha.configure do |config|
  config.spec_dir     = "spec/javascripts/mocha"
  config.spec_matcher = /_spec\.|_test\./
  # driver names a Capybara driver used for the run task (try :poltergeist, after installing PhantomJS).
  # Without this directive, konacha will try to run this in Firefox
  require 'capybara/poltergeist'
  config.driver = :poltergeist

dmichael

# Create the wrapped function with 5 retries
command = @commands.retry('connectToProvider', attempts: 5)

# Create the wrapped function with unlimited retries
command = @commands.retry('connectToProvider')

# Essentially a promise is returned with another method attached
promise = command('yahoo', username: 'x', password: 'y')

# Some bad shit happens

dmichael

require 'squeel'

class RealtimeMessage < ActiveRecord::Base
  # These attributes are "mass-assignable", that is, they come in directly from the serial port as XML
  attr_accessible :src, :dsb, :time, :tmprF, :sensor_num, :sensor_type, :radio_id, :ch1_watts, :ch2_watts, :ch3_watts, :mac_address
  
  def self.this_month 
    now            = Time.now
    first_of_month = Date.civil(now.year, now.month, 1)
    last_of_month  = Date.civil(now.year, now.month, -1)

dmichael

# Thanks Seth Darlington!

price_per_kwh = 0.1248 # dollars
  
query = RealtimeMessage.this_month

previous = query.first
usage = query[1..query.count].reduce(0) do |sum, message|
  elapsed_seconds = (message.created_at - previous.created_at)
  elapsed_hours   = elapsed_seconds/3600

dmichael

function g = sigmoid(z)
  g = sigmoid = 1.0 ./ (1.0 + exp(-z));
end

function g = sigmoidGradient(z)
  g = sigmoid(z) .* (1 - sigmoid(z));
end

dmichael

regularization = lambda/(2*m) * ( ...
  sum(sum((Theta1(:,2:end)).^2)) +  ...
  sum(sum((Theta2(:,2:end)).^2)) ...
);

J = (1/m) * sum( ...
  sum((-y .* log(h))-((1-y) .* log(1-h))) ...
);

J = J + regularization;

dmichael

J = (1/m) * sum( ...
  sum((-y .* log(h))-((1-y) .* log(1-h))) ...
);

dmichael

m  = size(X, 1);
a1 = [ones(m, 1) X];  % Add the biases to the input matrix

z2 = a1 * Theta1';
a2 = sigmoid(z2);

n  = size(a2, 1);
a2 = [ones(n,1) a2];  % Add the biases

z3 = a2 * Theta2';

dmichael

function g = sigmoid(z)
  % Compute the sigmoid of z. 
  % z can be a matrix, vector or scalar.
  g = 1.0 ./ (1.0 + exp(-z));
  % 1./ (1 + e.^-z); <-- this is equivalent
end

hypothesis = sigmoid(X*theta);

dmichael

function [J, grad] = costFunction(theta, X, y)
  % Logistic Regression cost function and partial derivative (gradient)
   
  m = length(y); % number of training examples

  cost = (y' * log(h)) + ((1-y')*log(1-h));

  J = -1/m * cost;

  grad = 1/m * (X' * (h - y));