-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathindex.html
More file actions
265 lines (249 loc) · 13 KB
/
Copy pathindex.html
File metadata and controls
265 lines (249 loc) · 13 KB
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
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<meta name="viewport" content="width=device-width, initial-scale=1">
<meta name="description" content="A promotional hub unifying PyOD, PyGDebias, and PyGIP into one interactive toolkit narrative.">
<title>IGNITE Toolkit Hub</title>
<link rel="preconnect" href="https://img.shields.io">
<link rel="preconnect" href="https://static.pepy.tech">
<link rel="stylesheet" href="styles.css">
</head>
<body>
<header class="hero" id="top">
<nav class="nav" aria-label="Primary navigation">
<a class="brand" href="#top" aria-label="IGNITE Toolkit Hub home">
<span class="brand-mark">I</span>
<span>IGNITE Toolkits</span>
</a>
<div class="nav-links">
<a href="#toolkits">Toolkits</a>
<a href="#pipeline">Platform</a>
<a href="#assistant">Ask</a>
<a href="#install">Install</a>
</div>
</nav>
<section class="hero-grid">
<div class="hero-copy">
<p class="eyebrow">Unified Trustworthy ML Toolkit Ecosystem</p>
<h1>Three mature Python libraries, one integrated story for trustworthy AI.</h1>
<p class="lead">
PyOD, PyGDebias, and PyGIP cover anomaly detection, fair graph learning, and graph
intellectual property protection. This hub frames them as a shared toolkit layer with
measurable adoption, ready-to-run APIs, and interactive pathways for new users.
</p>
<div class="hero-actions">
<a class="button primary" href="#assistant">Ask how to use it</a>
<a class="button secondary" href="#toolkits">Compare toolkits</a>
</div>
</div>
<div class="hero-panel" aria-label="Impact dashboard">
<div class="impact-card large">
<span class="metric">40M+</span>
<span class="label">PyOD public download signal</span>
</div>
<div class="impact-card">
<span class="metric">60+</span>
<span class="label">PyOD detectors</span>
</div>
<div class="impact-card">
<span class="metric">26</span>
<span class="label">PyGDebias graph datasets</span>
</div>
<div class="impact-card">
<span class="metric">13</span>
<span class="label">PyGDebias fair graph algorithms</span>
</div>
</div>
</section>
</header>
<main>
<section class="section narrative" aria-labelledby="narrative-title">
<div>
<p class="eyebrow">Proposal-Aligned Aim</p>
<h2 id="narrative-title">From separate packages to a single adoption narrative.</h2>
</div>
<div class="narrative-text">
<p>
The page positions the three libraries as a coherent backend and frontend ecosystem:
standardized installation, comparable metadata, unified entry points, and a guided
interface that answers practical user questions without requiring a separate user-study
section.
</p>
</div>
</section>
<section class="section" id="toolkits" aria-labelledby="toolkits-title">
<div class="section-head">
<p class="eyebrow">Adoption Signals</p>
<h2 id="toolkits-title">Popularity badges that make impact visible at a glance.</h2>
</div>
<div class="toolkit-grid">
<article class="toolkit-card" data-toolkit="pyod">
<div class="card-top">
<div>
<p class="kicker">Outlier Detection</p>
<h3>PyOD</h3>
</div>
<a class="repo-link" href="https://github.com/yzhao062/pyod" target="_blank" rel="noreferrer">GitHub</a>
</div>
<p class="card-copy">
A battle-tested Python library for scalable anomaly and outlier detection, with
classical, deep, LLM-assisted, and benchmark-backed model selection workflows.
</p>
<div class="badge-row" aria-label="PyOD popularity badges">
<img alt="PyOD GitHub stars" src="https://img.shields.io/github/stars/yzhao062/pyod?style=for-the-badge&logo=github&label=Stars">
<img alt="PyOD GitHub forks" src="https://img.shields.io/github/forks/yzhao062/pyod?style=for-the-badge&logo=github&label=Forks">
<img alt="PyOD PyPI downloads" src="https://static.pepy.tech/badge/pyod">
<img alt="PyOD PyPI version" src="https://img.shields.io/pypi/v/pyod?style=for-the-badge&logo=pypi&label=PyPI">
</div>
<ul class="feature-list">
<li>60+ detection algorithms under a consistent API.</li>
<li>Adoption-heavy anchor for the overall ecosystem narrative.</li>
<li>Works as the anomaly-detection backend service in the platform.</li>
</ul>
</article>
<article class="toolkit-card" data-toolkit="pygdebias">
<div class="card-top">
<div>
<p class="kicker">Fair Graph Learning</p>
<h3>PyGDebias</h3>
</div>
<a class="repo-link" href="https://github.com/yushundong/PyGDebias" target="_blank" rel="noreferrer">GitHub</a>
</div>
<p class="card-copy">
A PyTorch Geometric-based benchmark and toolkit for graph debiasing, bundling
datasets, fairness-aware algorithms, and unified evaluation flows.
</p>
<div class="badge-row" aria-label="PyGDebias popularity badges">
<img alt="PyGDebias GitHub stars" src="https://img.shields.io/github/stars/yushundong/PyGDebias?style=for-the-badge&logo=github&label=Stars">
<img alt="PyGDebias GitHub forks" src="https://img.shields.io/github/forks/yushundong/PyGDebias?style=for-the-badge&logo=github&label=Forks">
<img alt="PyGDebias PyPI downloads" src="https://static.pepy.tech/badge/pygdebias">
<img alt="PyGDebias PyPI version" src="https://img.shields.io/pypi/v/pygdebias?style=for-the-badge&logo=pypi&label=PyPI">
</div>
<ul class="feature-list">
<li>26 graph datasets for fairness-oriented experiments.</li>
<li>13 implemented algorithms spanning preprocessing, in-processing, and postprocessing.</li>
<li>Provides the graph-fairness service in the unified toolkit stack.</li>
</ul>
</article>
<article class="toolkit-card" data-toolkit="pygip">
<div class="card-top">
<div>
<p class="kicker">Graph IP Protection</p>
<h3>PyGIP</h3>
</div>
<a class="repo-link" href="https://github.com/LabRAI/PyGIP" target="_blank" rel="noreferrer">GitHub</a>
</div>
<p class="card-copy">
A modular framework for graph model intellectual property protection, including
model extraction attacks, defense methods, datasets, and graph neural network models.
</p>
<div class="badge-row" aria-label="PyGIP popularity badges">
<img alt="PyGIP GitHub stars" src="https://img.shields.io/github/stars/LabRAI/PyGIP?style=for-the-badge&logo=github&label=Stars">
<img alt="PyGIP GitHub forks" src="https://img.shields.io/github/forks/LabRAI/PyGIP?style=for-the-badge&logo=github&label=Forks">
<img alt="PyGIP PyPI downloads" src="https://static.pepy.tech/badge/pygip">
<img alt="PyGIP PyPI version" src="https://img.shields.io/pypi/v/pygip?style=for-the-badge&logo=pypi&label=PyPI">
</div>
<ul class="feature-list">
<li>Attack and defense modules for graph model IP risk analysis.</li>
<li>Experiment-ready structure for GNN models and datasets.</li>
<li>Adds the model-protection service to the integrated platform.</li>
</ul>
</article>
</div>
</section>
<section class="section platform" id="pipeline" aria-labelledby="pipeline-title">
<div class="section-head">
<p class="eyebrow">Unified Architecture</p>
<h2 id="pipeline-title">A platform layer that makes three libraries feel like one product.</h2>
</div>
<div class="pipeline">
<div class="pipe-step">
<span>01</span>
<h3>Shared Registry</h3>
<p>Expose each algorithm, dataset, metric, and tutorial through a searchable metadata layer.</p>
</div>
<div class="pipe-step">
<span>02</span>
<h3>Microservice APIs</h3>
<p>Wrap PyOD, PyGDebias, and PyGIP behind consistent endpoints for run, compare, and export.</p>
</div>
<div class="pipe-step">
<span>03</span>
<h3>Visual Builder</h3>
<p>Let users assemble detection, debiasing, and protection workflows without reading every API first.</p>
</div>
<div class="pipe-step">
<span>04</span>
<h3>LLM Assistant</h3>
<p>Answer “which toolkit do I need?” and “how do I install it?” with grounded, copyable recipes.</p>
</div>
</div>
</section>
<section class="section assistant-section" id="assistant" aria-labelledby="assistant-title">
<div class="assistant-copy">
<p class="eyebrow">Interactive Demo</p>
<h2 id="assistant-title">Ask practical questions and get a focused answer.</h2>
<p>
Use the presets or type a phrase such as “fair graph learning”, “outlier detection”,
“model extraction”, “install PyGDebias”, or “compare all”.
</p>
<div class="question-chips" aria-label="Suggested questions">
<button type="button" data-question="Which toolkit should I use for anomaly detection?">Anomaly detection</button>
<button type="button" data-question="How do I install PyGDebias?">Install PyGDebias</button>
<button type="button" data-question="How can PyGIP help with model extraction risk?">Graph IP risk</button>
<button type="button" data-question="Compare all three toolkits.">Compare all</button>
</div>
</div>
<div class="assistant-panel">
<label for="questionInput">Question</label>
<div class="ask-row">
<input id="questionInput" type="search" placeholder="Ask about install, usage, or toolkit selection">
<button id="askButton" type="button">Ask</button>
</div>
<div id="answerBox" class="answer-box" aria-live="polite">
Choose a question to generate a short, presentation-ready answer.
</div>
</div>
</section>
<section class="section install" id="install" aria-labelledby="install-title">
<div class="section-head">
<p class="eyebrow">Quick Start</p>
<h2 id="install-title">Copyable install recipes for each audience.</h2>
</div>
<div class="install-tabs" role="tablist" aria-label="Install commands">
<button class="tab is-active" type="button" data-target="pyod-code">PyOD</button>
<button class="tab" type="button" data-target="pygdebias-code">PyGDebias</button>
<button class="tab" type="button" data-target="pygip-code">PyGIP</button>
</div>
<div class="code-shell">
<pre id="pyod-code" class="code-panel is-active"><code>pip install pyod
from pyod.models.iforest import IForest
clf = IForest()
clf.fit(X_train)
scores = clf.decision_scores_</code></pre>
<pre id="pygdebias-code" class="code-panel"><code>pip install pygdebias
from pygdebias.datasets import Bail
dataset = Bail()
adj, features, labels, idx_train, idx_val, idx_test, sens = dataset.load_data()</code></pre>
<pre id="pygip-code" class="code-panel"><code>pip install PyGIP
# Use PyGIP's modular attack/defense components to evaluate
# graph model IP risk under model extraction settings.</code></pre>
<button class="copy-button" id="copyButton" type="button">Copy command</button>
</div>
</section>
<section class="section sources" aria-labelledby="sources-title">
<h2 id="sources-title">Source Links</h2>
<div class="source-links">
<a href="https://github.com/yzhao062/pyod" target="_blank" rel="noreferrer">PyOD GitHub</a>
<a href="https://pypi.org/project/pyod/" target="_blank" rel="noreferrer">PyOD PyPI</a>
<a href="https://github.com/yushundong/PyGDebias" target="_blank" rel="noreferrer">PyGDebias GitHub</a>
<a href="https://pypi.org/project/pygdebias/" target="_blank" rel="noreferrer">PyGDebias PyPI</a>
<a href="https://github.com/LabRAI/PyGIP" target="_blank" rel="noreferrer">PyGIP GitHub</a>
<a href="https://pypi.org/project/pygip/" target="_blank" rel="noreferrer">PyGIP PyPI</a>
</div>
</section>
</main>
<script src="app.js" type="module"></script>
</body>
</html>