-
Notifications
You must be signed in to change notification settings - Fork 1
/
index.html
89 lines (85 loc) · 3.05 KB
/
index.html
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
<html>
<head>
<meta charset="utf-8">
<title>Active Revision Sampling - Supplementary Visualizations</title>
</head>
<body>
<h1>Active Revision Sampling - Supplementary Visualizations</h1>
<h2>Change point analysis</h2>
<table>
<tr>
<td>Subject System</td><td>Visualization</td>
</tr>
<tr>
<td>xz</td><td>
<a href="xz.html">xz (47 variants)</a>
</td>
</tr>
<tr>
<td>lrzip</td><td>
<a href="lrzip.html">lrzip (47 variants)</a>
</td>
</tr>
<tr>
<td>ultrajson</td><td>
<a href="ultrajson.html">ultrajson (4 methods)</a>
</td>
</tr>
<tr>
<td>pillow</td><td>
<a href="pillow.html">pillow (59 methods)</a>
</td>
</tr>
<tr>
<td>numpy</td><td>
<a href="numpy.html">numpy (204 methods)</a>
</td>
</tr>
<tr>
<td>scipy</td><td>
<a href="scipy.html">scipy (174 methods)</a>
</td>
</tr>
</table>
<h2>Visualizations of Active Learning</h2>
The following interactive visualizations present the progress of active learning using Gaussian Process regression with the five different kernels used (Radias Basis Function Kernel (RBF), Rational Quadratic (RatQuad), Brownian Kernel (Brownian), and the two different parametrizations of the Matérn kernel (Matern32 and Matern52)). Each of the five subplots represents a different kernel and each data point represents an individual variant (for the subject systems xz and lrzip) or microbenchmark (for ultrajson, pillow, numpy, and scipy). <br \>
We provide the visualizations on separate websites, which are best be viewed offline since the visualizations range from 3 MB to over 100 MB in size. Hence, please, find the compressed visualizations below.
<table>
<tr>
<td>Subject System</td><td>Visualization</td>
</tr>
<tr>
<td>xz</td><td>
<a href="animations/rq1_xz.html.zip">zipped HTML (1.5 MiB, 5.3 MiB when extracted)</a>
</td>
</tr>
<tr>
<td>lrzip</td><td>
<a href="animations/rq1_lrzip.html.zip">zipped HTML (1.5 MiB, 4.9 MiB when extracted)</a>
</td>
</tr>
<tr>
<td>ultrajson</td><td>
<a href="animations/rq1_ultrajson.html.zip">zipped HTML (875.6 KiB, 3.1 MiB when extracted)</a>
</td>
</tr>
<tr>
<td>pillow</td><td>
<a href="animations/rq1_pillow.html.zip">zipped HTML (6 MiB, 43.1 MiB when extracted)</a>
</td>
</tr>
<tr>
<td>numpy</td><td>
<a href="animations/rq1_numpy.html.zip">zipped HTML (12.6 MiB, 111.7 MiB when extracted)</a>
</td>
</tr>
<tr>
<td>scipy</td><td>
<a href="animations/rq1_scipy.html.zip">zipped HTML (8.1 MiB, 65.4 MiB when extracted)</a>
</td>
</tr>
</table>
<h2>Implementation</h2>
The implemenation of our active learner and the change point estimation from RQ3 can be accessed <a href="implementation">here</a>
</body>
</html>