-
Notifications
You must be signed in to change notification settings - Fork 1
/
Copy pathindex.html
182 lines (177 loc) · 7.4 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
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
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
<!DOCTYPE html>
<html>
<head>
<meta charset="utf-8" />
<meta name="viewport" content="width=device-width" />
<title>Fit XGBoost models online | StatSim Fit</title>
<link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/materialize/1.0.0/css/materialize.min.css">
<link type="text/css" rel="stylesheet" href="https://statsim.com/port/css/port.css" media="screen"/>
<link rel="icon" type="image/png" href="https://statsim.com/app/images/favicon-32x32.png" sizes="32x32">
<link rel="icon" type="image/png" href="https://statsim.com/app/images/favicon-16x16.png" sizes="16x16">
<link type="text/css" rel="stylesheet" href="https://statsim.com/assets/common.css" media="screen"/>
<style>
body {
background-color: #ffffff;
background: #f5f5f5;
}
a { color: #7100E3 }
h4 { font-weight: 700; }
.file-field .btn { background: #BBB }
.file-field .btn:hover { background: #AAA }
.btn, .port-btn { background: #7100E3 }
.btn:hover, .port-btn:hover { background: #5700AD }
.grey-bar {
background: white;
box-shadow: 0px 1px 25px #ebebeb;
}
ol {
background:#f5f5f5;
padding: 20px 0 20px 25px;
border-radius: 10px;
margin: 50px 0 20px 0;
font-size: 16px;
list-style: none;
counter-reset: custom-counter;
}
ol li {
counter-increment: custom-counter;
}
ol li::before {
content: counter(custom-counter) ". ";
color: #BBB;
font-weight: bold;
margin-right: 5px;
}
.spinner-green, .spinner-green-only { border-color: #3030B7 }
#outputs:empty { display: none; }
</style>
<!-- Global site tag (gtag.js) - Google Analytics -->
<script async src="https://www.googletagmanager.com/gtag/js?id=UA-7770107-2"></script>
<script>
window.dataLayer = window.dataLayer || [];
function gtag(){dataLayer.push(arguments);}
gtag('js', new Date());
gtag('config', 'UA-7770107-2');
</script>
</head>
<body>
<div class="status-bar grey-bar">
<div class="container">
<div class="row">
<div class="col s12" style="font-size: 14px;">
<div id="menu"></div>
<a href="https://statsim.com/">StatSim</a> → <b>Fit</b>
<span class="version">Version: 0.0.2</span>
</div>
</div>
</div>
</div>
<div class="container">
<div class="row">
<div id="port-container"></div>
</div>
</div>
<div id="description" class="grey-bar">
<div class="container">
<div class="row">
<div class="col m12">
<h1>Fit XGBoost models online</h1>
<h2>Easily fit a machine learning model and predict new values</h2>
<ol>
<li>Select a training dataset in the <b>CSV</b> format</li>
<li>Select a dataset for prediction</li>
<li>Choose a target variable (variable you want to predict)</li>
<li>Customize <b>XGBoost</b> model</li>
<li>Click <b>Run</b></li>
</ol>
</div>
</div>
<div class="row features">
<div class="col m3 feature">
<h3>
The best model for tabular data
</h3>
<p>
XGBoost is probably the best model for tabular data with independent observations (not time-series). It's based on an ensemble of decision trees and won a lot of public competitions. XGBoost supports classification, regression, count, and survival modes.
</p>
</div>
<div class="col m3 feature">
<h3>
Cross-validation for small datasets
</h3>
<p>
In many cases, a dataset is too small to evaluate a model performance correctly. A standard practice of splitting data on training and test sets gives unreliable results. Cross-validation helps with this problem by running multiple splits iteratively and calculating a more robust score based on those runs.
</p>
</div>
<div class="col m3 feature">
<h3>
Download a model and its predictions
</h3>
<p>
After a model is fit and made predictions, you can download a resulting CSV file for further usage. It includes the same input columns together with the predicted target column. You can also download the fitted model in JSON format and charts with XGBoost performance on this dataset as SVGs.
</p>
</div>
<div class="col m3 feature">
<h3>
Client-side processing
</h3>
<p>
XGBoost is a native C++ library ported to WebAssembly and wrapped in this web application. When you load a dataset, it's processed using a virtual machine running in your browser. That means no single byte of your data is leaving a device that runs this web app.
</p>
</div>
</div>
<div class="row">
<div class="col m12">
<small>
If you enjoyed the app, star us on GitHub. To report errors, create an issue.<br>
</small>
<p>
<a class="github-button" href="https://github.com/statsim/fit" data-icon="octicon-star" data-show-count="true" aria-label="Star statsim/fit on GitHub">Star</a>
<a class="github-button" href="https://github.com/statsim/fit/issues" data-icon="octicon-issue-opened" data-show-count="true" aria-label="Issue statsim/fit on GitHub">Issue</a>
</p>
</div>
</div>
</div>
</div>
<script async defer src="https://buttons.github.io/buttons.js"></script>
<script src="https://cdnjs.cloudflare.com/ajax/libs/materialize/1.0.0/js/materialize.min.js"></script>
<script src="https://statsim.com/port/dist/port.js"></script>
<script src="https://statsim.com/assets/common.js"></script>
<script>
var port = new Port({
portContainer: document.getElementById('port-container'),
schema: {
"model": {
"name": "Process",
"method": "run",
"type": "class",
"url": "process.js",
"worker": true,
},
"design": {
"layout": "blocks",
"colors": "light"
},
"inputs": [
{ "type": "file", "name": "Train set", "reactive": true },
{ "type": "file", "name": "Test set"},
{ "type": "select", "name": "Target variable" },
{ "type": "select", "name": "Model type", "options": ["XGBoost Classifier", "XGBoost Regressor", "XGBoost Survival", "XGBoost Count"], "default": "XGBoost Classifier" },
{ "type": "int", "name": "Max depth", "default": 5, "min": 0 },
{ "type": "int", "name": "Iterations", "default": 100, "min": 0 },
{ "type": "checkbox", "name": "Cross-validation", "default": false },
],
"outputs": [
{ "type": "string", "name": "Train set" },
{ "type": "string", "name": "Train set (CV)" },
{ "type": "string", "name": "Test set" },
{ "type": "svg", "name": "CV" },
{ "type": "svg", "name": "Plot" },
{ "type": "file", "name": "Model" },
{ "type": "file", "name": "Prediction" },
]
}
})
</script>
</body>
</html>