-
Notifications
You must be signed in to change notification settings - Fork 1
/
sketch.js
133 lines (118 loc) · 3.33 KB
/
sketch.js
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
let x_value = [];
let y_value = [];
var polynomialType = "Linear";
var learningType = "SGD";
let m, b, a, c, d;
const learningRate = 0.09;
var optimizer = tf.train.sgd(learningRate);
let dragging = false;
function changeCanvas() {
clear();
polynomialType = document.getElementById("polynomialType").value;
learningType = document.getElementById("learningType").value;
if(learningType == "SGD") {
optimizer = tf.train.sgd(learningRate);
} else if(learningType == "Adam") {
optimizer = tf.train.adam(learningRate);
}
x_value = [];
y_value = [];
console.log(polynomialType);
console.log(learningType);
}
function setup() {
// put setup code here
var cnv = createCanvas(windowWidth - 80, windowHeight - 130);
var x = (windowWidth - width) / 2;
cnv.position(x);
cnv.parent("p5Canvas");
cnv.mousePressed(canvasMousePressed);
cnv.mouseReleased(canvasMouseReleased);
m = tf.variable(tf.scalar(random(-1, 1)));
b = tf.variable(tf.scalar(random(-1, 1)));
a = tf.variable(tf.scalar(random(-1, 1)));
c = tf.variable(tf.scalar(random(-1, 1)));
d = tf.variable(tf.scalar(random(-1, 1)));
}
function loss_func(pred, labels) {
return pred.sub(labels).square().mean();
}
function predict(x_value) {
const xs = tf.tensor1d(x_value);
if(polynomialType == "Linear") {
// y = mx + b
return xs.mul(a).add(b);
} else if (polynomialType == "Binomial") {
// y = ax^2 + bx + c
return xs.square().mul(a).add(b.mul(xs)).add(c);
} else if (polynomialType == "Trinomial") {
// y = ax^3 + bx^2 + cx + d
return xs.pow(tf.scalar(3)).mul(a)
.add(b.square().mul(xs))
.add(c.mul(xs))
.add(d);
}
}
function windowResized() {
resizeCanvas(windowWidth - 80, windowHeight - 80);
}
function canvasMousePressed() {
dragging = true;
}
function canvasMouseReleased() {
dragging = false;
}
function draw() {
background(51);
if(dragging) {
let x = map(mouseX, 0, width, -1, 1);
let y = map(mouseY, 0, height, 1, -1);
x_value.push(x);
y_value.push(y);
} else {
tf.tidy(() => {
if (x_value.length > 0) {
const ys = tf.tensor1d(y_value);
optimizer.minimize(() => loss_func(predict(x_value), ys));
}
});
}
// for pointing pixels on the screen
stroke(255);
strokeWeight(13);
for(let i = 0; i <= x_value.length; i++) {
let px = map(x_value[i], -1, 1, 0, width);
let py = map(y_value[i], -1, 1, height, 0);
point(px, py);
}
if(polynomialType == "Linear") {
let xs = [-1, 1];
const ys = tf.tidy(() => predict(xs));
let x1 = map(xs[0], -1, 1, 0, width);
let x2 = map(xs[1], -1, 1, 0, width);
let liney = ys.dataSync();
ys.dispose();
let y1 = map(liney[0], -1, 1, height, 0);
let y2 = map(liney[1], -1, 1, height, 0);
strokeWeight(2);
line(x1, y1, x2, y2);
} else if(polynomialType == "Binomial" || polynomialType == "Trinomial") {
const curveXs = [];
for(let i = -1; i <= 1.01; i += 0.02) {
curveXs.push(i);
}
const ys = tf.tidy(() => predict(curveXs));
let curveYs = ys.dataSync();
ys.dispose();
beginShape();
noFill();
stroke(255);
strokeWeight(2);
for (let i = 0; i < curveXs.length; i++) {
let xs = map(curveXs[i], -1, 1, 0, width);
let ys = map(curveYs[i], -1, 1, height, 0);
vertex(xs, ys);
}
endShape();
}
}