Evan Davis eridgd

## fc_conv_equivalence.py
import torch
import torch.nn as nn

lin = nn.Linear(9, 1, bias=False)
lin.weight.data = torch.ones_like(lin.weight)

x = torch.randn(1,1,3,3)

print(lin(x.flatten()))
# tensor([-1.7673], grad_fn=<SqueezeBackward3>)

## helloworld.js
var serialport = require('node-serialport')

var sp = new serialport.SerialPort("/dev/ttyO3", {
  parser: serialport.parsers.raw,
  baud: 9600
})

sp.on('data', function(chunk) {
  console.log(chunk.toString('hex'), chunk.toString(), chunk)
})

## cem.md

      
              2 files
            
          
              0 forks
            
          
                0 comments
              
            
              0 stars
            
          
                eridgd
                / cem.md
            
            
              Created
              March 1, 2018 23:15
                — forked from kashif/cem.md
            
              
                Cross Entropy Method
              
          
    Cross Entropy Method

How do we solve  for the policy optimization problem which is to maximize the total reward given some parametrized policy?
Discounted future reward

To begin with, for an episode the total reward is the sum of all the rewards. If our environment is stochastic, we can never be sure if we will get the same rewards the next time we perform the same actions. Thus the more we go into the future the more the total future reward may diverge. So for that reason it is common to use the discounted future reward where the parameter discount is called the discount factor and is between 0 and 1.
A good strategy for an agent would be to always choose an action that maximizes the (discounted) future reward. In other words we want to maximize the expected reward per episode.

  
## depth_to_space_viz.py
def _phase_shift(I, r):
    bsize, a, b, c = I.get_shape().as_list()
    X = tf.reshape(I, (bsize, a, b, r, r))
    X = tf.transpose(X, (0, 1, 2, 4, 3))  # bsize, a, b, 1, 1
    X = tf.split(1, a, X)  # a, [bsize, b, r, r]
    X = tf.concat(2, [tf.squeeze(x, [1]) for x in X])  # bsize, b, a*r, r
    X = tf.split(1, b, X)  # b, [bsize, a*r, r]
    X = tf.concat(2, [tf.squeeze(x, [1]) for x in X])  # bsize, a*r, b*r
    return tf.reshape(X, (bsize, a*r, b*r, 1))

## depth_to_space_viz.py
def _phase_shift(I, r):
    bsize, a, b, c = I.get_shape().as_list()
    X = tf.reshape(I, (bsize, a, b, r, r))
    X = tf.transpose(X, (0, 1, 2, 4, 3))  # bsize, a, b, 1, 1
    X = tf.split(1, a, X)  # a, [bsize, b, r, r]
    X = tf.concat(2, [tf.squeeze(x, [1]) for x in X])  # bsize, b, a*r, r
    X = tf.split(1, b, X)  # b, [bsize, a*r, r]
    X = tf.concat(2, [tf.squeeze(x, [1]) for x in X])  # bsize, a*r, b*r
    return tf.reshape(X, (bsize, a*r, b*r, 1))

## depth_to_space_viz.py
def _phase_shift(I, r):
    bsize, a, b, c = I.get_shape().as_list()
    X = tf.reshape(I, (bsize, a, b, r, r))
    X = tf.transpose(X, (0, 1, 2, 4, 3))  # bsize, a, b, 1, 1
    X = tf.split(1, a, X)  # a, [bsize, b, r, r]
    X = tf.concat(2, [tf.squeeze(x, [1]) for x in X])  # bsize, b, a*r, r
    X = tf.split(1, b, X)  # b, [bsize, a*r, r]
    X = tf.concat(2, [tf.squeeze(x, [1]) for x in X])  # bsize, a*r, b*r
    return tf.reshape(X, (bsize, a*r, b*r, 1))

## data.csv

          
            name
            region
            rating
            numReviews
            price

            
              1792 Ridgemont Reserve 
              Bourbon
              84
              1
              25

            
              Aberfeldy 12
              Highland
              73.5
              4
              42.6666666667

            
              Aberfeldy 1998 Signatory Vintage
              Highland
              74
              2
              40

            
              Aberlour 10
              Speyside
              81.5
              6
              46

            
              Aberlour 10 
              Speyside
              73
              1
              25

            
              Aberlour 11 Duncan Taylor NC2
              Speyside
              87.25
              4
              60

            
              Aberlour 12 Double Cask Matured
              Speyside
              85.4285714286
              14
              38.3333333333

            
              Aberlour 12 Non Chill Filtered
              Speyside
              93.375
              8
              56.5

            
              Aberlour 16
              Speyside
              88.75
              4
              54.3333333333

## index.html
<!DOCTYPE html>
<meta charset="utf-8">
<meta name="viewport" width="device-width" content="target-densitydpi=device-dpi" />
<title>Whisk(e)y Reviews by Rating and Price</title>
<style>
    body {
        font: 10px sans-serif;
    }
    .axis path, .axis line {
        fill: none;

## index.html
<!DOCTYPE html>
<meta charset="utf-8">
<style>
    body {
        font: 10px sans-serif;
    }
    .axis path, .axis line {
        fill: none;
        stroke: #000;
        shape-rendering: crispEdges;

## index.html
<!DOCTYPE html>
<meta charset="utf-8">
<style>
    body {
        font: 10px sans-serif;
    }
    .axis path, .axis line {
        fill: none;
        stroke: #000;
        shape-rendering: crispEdges;
	import torch
	import torch.nn as nn

	lin = nn.Linear(9, 1, bias=False)
	lin.weight.data = torch.ones_like(lin.weight)

	x = torch.randn(1,1,3,3)

	print(lin(x.flatten()))
	# tensor([-1.7673], grad_fn=<SqueezeBackward3>)
	var serialport = require('node-serialport')

	var sp = new serialport.SerialPort("/dev/ttyO3", {
	parser: serialport.parsers.raw,
	baud: 9600
	})

	sp.on('data', function(chunk) {
	console.log(chunk.toString('hex'), chunk.toString(), chunk)
	})
	def _phase_shift(I, r):
	bsize, a, b, c = I.get_shape().as_list()
	X = tf.reshape(I, (bsize, a, b, r, r))
	X = tf.transpose(X, (0, 1, 2, 4, 3)) # bsize, a, b, 1, 1
	X = tf.split(1, a, X) # a, [bsize, b, r, r]
	X = tf.concat(2, [tf.squeeze(x, [1]) for x in X]) # bsize, b, a*r, r
	X = tf.split(1, b, X) # b, [bsize, a*r, r]
	X = tf.concat(2, [tf.squeeze(x, [1]) for x in X]) # bsize, ar, br
	return tf.reshape(X, (bsize, ar, br, 1))
name	region	rating	numReviews	price
1792 Ridgemont Reserve	Bourbon	84	1	25
Aberfeldy 12	Highland	73.5	4	42.6666666667
Aberfeldy 1998 Signatory Vintage	Highland	74	2	40
Aberlour 10	Speyside	81.5	6	46
Aberlour 10	Speyside	73	1	25
Aberlour 11 Duncan Taylor NC2	Speyside	87.25	4	60
Aberlour 12 Double Cask Matured	Speyside	85.4285714286	14	38.3333333333
Aberlour 12 Non Chill Filtered	Speyside	93.375	8	56.5
Aberlour 16	Speyside	88.75	4	54.3333333333
	<!DOCTYPE html>
	<meta charset="utf-8">
	<meta name="viewport" width="device-width" content="target-densitydpi=device-dpi" />
	<title>Whisk(e)y Reviews by Rating and Price</title>
	<style>
	body {
	font: 10px sans-serif;
	}
	.axis path, .axis line {
	fill: none;