Skip to content

Commit 430dd5e

Browse files
tiskwtiskw
committed
Merge branch 'bugfix/doc'
2 parents 49ad556 + 4d70ee7 commit 430dd5e

22 files changed

+762
-783
lines changed

‎README.md‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -26,7 +26,7 @@ machine [3, 4] and Gaussian process model [5]. Features of this module are:
2626
demonstrate that RFF is useful for actual machine learning tasks,
2727
- **GPU support**: Some classes in the `rfflearn` module provides both GPU training and inference
2828
for faster computation,
29-
- **Wrapper to the other library**: Interface to [optuna](https://optuna.org/) and
29+
- **Wrapper to the other library**: Interface to [Optuna](https://optuna.org/) and
3030
[SHAP](https://github.com/slundberg/shap) are provided for easier hyperparameter tuning and
3131
feature importance analysis.
3232

‎SETUP.md‎

Lines changed: 0 additions & 88 deletions
This file was deleted.

‎docs/api_reference.html‎

Lines changed: 86 additions & 35 deletions
Original file line numberDiff line numberDiff line change
@@ -20,8 +20,8 @@
2020
<link rel="stylesheet" href="https://fonts.googleapis.com/css2?family=Montserrat&display=swap" media="print" onload="this.media='all'">
2121

2222
<!-- Load Highlight JS -->
23-
<link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/highlight.js/9.15.10/styles/atelier-forest-light.min.css">
24-
<script src="https://cdnjs.cloudflare.com/ajax/libs/highlight.js/9.15.10/highlight.min.js"></script>
23+
<link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/highlight.js/11.10.0/styles/stackoverflow-light.min.css">
24+
<script src="https://cdnjs.cloudflare.com/ajax/libs/highlight.js/11.10.0/highlight.min.js"></script>
2525

2626
<!-- Bootstrap core CSS -->
2727
<link rel="stylesheet" href="./css/bootstrap.min.css">
@@ -41,44 +41,41 @@
4141

4242
<main>
4343

44-
<!-- -->
44+
<!-- Main sections -->
4545
<section class="container py-3">
4646
<h1 class="py-2 border-bottom" id="setting_up">API Reference</h1>
4747
<p>The RFFLearn library consists of the several sub modules. <!--
4848
-->Click the module name to see the details of each module.</p>
4949

5050
<div class="row align-items-center justify-content-center"><div class="col-8"><table class="table table-hover">
51-
<thead><tr>
52-
<th scope="col">Sub module name</th>
53-
<th scope="col">Description</th>
51+
<thead class="head"><tr class="head">
52+
<th class="head" scope="col">Sub module name</th>
53+
<th class="head" scope="col">Description</th>
5454
</tr></thead>
5555
<tbody><tr>
5656
<td><a href="#rfflearn_cpu">rfflearn.cpu</a></td>
5757
<td>Regressors and classifiers of random Fourier features on CPU.</td>
5858
</tr><tr>
5959
<td><a href="#rfflearn_gpu">rfflearn.gpu</a></td>
6060
<td>Regressors and classifiers of random Fourier features on GPU.</td>
61-
</tr><tr>
62-
<td><a href="#rfflearn_explainer">rfflearn.explainer</a></td>
63-
<td>Classes and functions to get model explanation for RFF-based models.</td>
64-
</tr><tr>
65-
<td><a href="#rfflearn_tuner">rfflearn.tuner</a></td>
66-
<td>Classes and functions for automatic hyperparameter tuning.</td>
6761
</tr></tbody>
6862
</table></div></div>
6963
</section>
7064

7165
<section class="container py-3">
7266
<h2 class="py-2" id="rfflearn_cpu">rfflearn.cpu</h2>
73-
<p>Sub module for machine learning algorithm of random Fourier feature runnable on CPU. <!--
74-
-->This sub module contains machine learning algorithm (e.g. regressors, classifiers) which is designed to be run on CPU. <!--
75-
-->Interfaces of classes and functions in this module have quite close interfaces with scikit-learn library. <!--
76-
-->Most (but not all) of the classes uses scikit-learn as a back end.</p>
77-
78-
<div class="row align-items-center justify-content-center"><div class="col-8"><table class="table table-hover">
79-
<thead><tr>
80-
<th scope="col">Class name</th>
81-
<th scope="col">Description</th>
67+
<p>This sub-module is a module for machine learning algorithms of random Fourier feature runnable on CPU. <!--
68+
-->This sub-module contains machine learning algorithms (e.g. regressors, classifiers) that are designed <!--
69+
-->to be run on CPU. The interfaces of classes and functions in this module are quite close to <!--
70+
-->the scikit-learn library. Most (but not all) of the classes use the scikit-learn as a backend.</p>
71+
<p>This sub-module contains three types of classes and functions: (1) ML models such as support vector machines <!--
72+
-->with random Fourier features, (2) hyperparameter tuners that are wrapper functions of Optuna, and (3) ML model <!--
73+
-->explainers that are wrapper functions of explainer functions in scikit-learn and SHAP.</p>
74+
75+
<div class="row align-items-center justify-content-center"><div class="col-8"><table class="table table-hover mt-2">
76+
<thead><tr class="heading">
77+
<th class="head" scope="col">ML class</th>
78+
<th class="head" scope="col">Description</th>
8279
</tr></thead>
8380
<tbody><tr>
8481
<td><a href="./api_reference_RFFCCA.html">rfflearn.cpu.RFFCCA</a></td>
@@ -101,21 +98,85 @@ <h2 class="py-2" id="rfflearn_cpu">rfflearn.cpu</h2>
10198
</tr><tr>
10299
<td><a href="./api_reference_RFFSVR.html">rfflearn.cpu.RFFSVR</a></td>
103100
<td>Support vector regression with random Fourier features.</td>
104-
</tr><tr>
105-
<td><a href="./api_reference_RFFBatchSVC.html">rfflearn.cpu.RFFBatchSVC</a></td>
106-
<td>Batch learning version of <code>rfflearn.cpu.RFFSVC</code>.</td>
107101
</tr><tr>
108102
<td><a href="#">rfflearn.cpu.ORF*</a></td>
109103
<td>ORF (orthogonal random features) version of estimators. For example, <code>rfflearn.cpu.ORFSVC</code> <!--
110104
-->is a support vector classifier with ORF. The arguments of constructor and member functions are the same as RFF version, <!--
111105
-->so please see the document of <code>rfflearn.cpu.RFF*</code> for the details of the usage of each class.</td>
112106
</tr><tr>
113107
<td><a href="#">rfflearn.cpu.QRF*</a></td>
114-
<td>QRF (quasi random features) version of estimators. For example, <code>rfflearn.cpu.QRFSVC</code> <!--
108+
<td>QRF (quasi-random features) version of estimators. For example, <code>rfflearn.cpu.QRFSVC</code> <!--
115109
-->is a support vector classifier with QRF. The arguments of constructor and member functions are the same as RFF version, <!--
116110
-->so please see the document of <code>rfflearn.cpu.RFF*</code> for the details of the usage of each class.</td>
117111
</tr></tbody>
118112
</table></div></div>
113+
114+
<div class="row align-items-center justify-content-center"><div class="col-8"><table class="table table-hover mt-3">
115+
<thead><tr class="heading">
116+
<th class="head" scope="col">Explainer function</th>
117+
<th class="head" scope="col">Description</th>
118+
</tr></thead>
119+
<tbody><tr>
120+
<td><a href="./api_reference_explainers.html#permutation_feature_importance">rfflearn.cpu.permutation_feature_importance</a></td>
121+
<td>Calculate permutation importance, and set the feature importance <!--
122+
-->(mean of permutation importance for each trial) as <code>model.feature_importances_</code>.</td>
123+
</tr><tr>
124+
<td><a href="./api_reference_explainers.html#permutation_plot">rfflearn.cpu.permutation_plot</a></td>
125+
<td>Visualize permutation importance as a box diagram.</td>
126+
</tr><tr>
127+
<td><a href="./api_reference_explainers.html#shap_feature_importance">rfflearn.cpu.shap_feature_importance</a></td>
128+
<td>Calculate SHAP values using shap library, and set the feature importance <!--
129+
-->(absolute of SHAP values) as <code>model.feature_importances_</code>.</td>
130+
</tr><tr>
131+
<td><a href="./api_reference_explainers.html#shap_plot">rfflearn.cpu.shap_plot</a></td>
132+
<td>Create a bar plot of SHAP values.</td>
133+
</tr></tbody>
134+
</table></div></div>
135+
136+
<div class="row align-items-center justify-content-center"><div class="col-8"><table class="table table-hover mt-3">
137+
<thead><tr class="heading">
138+
<th class="head" scope="col">Tuner function</th>
139+
<th class="head" scope="col">Description</th>
140+
</tr></thead>
141+
<tbody><tr>
142+
<td><a href="./api_reference_tuners.html">rfflearn.cpu.RFFRegressor_tuner</a></td>
143+
<td>Hyperparameter tuner for <code>RFFRegressor</code> models.</td>
144+
</tr><tr>
145+
<td><a href="./api_reference_tuners.html">rfflearn.cpu.ORFRegressor_tuner</a></td>
146+
<td>Hyperparameter tuner for <code>OFFRegressor</code> models.</td>
147+
</tr><tr>
148+
<td><a href="./api_reference_tuners.html">rfflearn.cpu.QRFRegressor_tuner</a></td>
149+
<td>Hyperparameter tuner for <code>QFFRegressor</code> models.</td>
150+
</tr><tr>
151+
<td><a href="./api_reference_tuners.html">rfflearn.cpu.RFFSVC_tuner</a></td>
152+
<td>Hyperparameter tuner for <code>RFFSVC</code> models.</td>
153+
</tr><tr>
154+
<td><a href="./api_reference_tuners.html">rfflearn.cpu.ORFSVC_tuner</a></td>
155+
<td>Hyperparameter tuner for <code>ORFSVC</code> models.</td>
156+
</tr><tr>
157+
<td><a href="./api_reference_tuners.html">rfflearn.cpu.QRFSVC_tuner</a></td>
158+
<td>Hyperparameter tuner for <code>QRFSVC</code> models.</td>
159+
</tr><tr>
160+
<td><a href="./api_reference_tuners.html">rfflearn.cpu.RFFGPC_tuner</a></td>
161+
<td>Hyperparameter tuner for <code>RFFGPC</code> models.</td>
162+
</tr><tr>
163+
<td><a href="./api_reference_tuners.html">rfflearn.cpu.ORFGPC_tuner</a></td>
164+
<td>Hyperparameter tuner for <code>ORFGPC</code> models.</td>
165+
</tr><tr>
166+
<td><a href="./api_reference_tuners.html">rfflearn.cpu.QRFGPC_tuner</a></td>
167+
<td>Hyperparameter tuner for <code>QRFGPC</code> models.</td>
168+
</tr><tr>
169+
<td><a href="./api_reference_tuners.html">rfflearn.cpu.RFFGPR_tuner</a></td>
170+
<td>Hyperparameter tuner for <code>RFFGPR</code> models.</td>
171+
</tr><tr>
172+
<td><a href="./api_reference_tuners.html">rfflearn.cpu.ORFGPR_tuner</a></td>
173+
<td>Hyperparameter tuner for <code>ORFGPR</code> models.</td>
174+
</tr><tr>
175+
<td><a href="./api_reference_tuners.html">rfflearn.cpu.QRFGPR_tuner</a></td>
176+
<td>Hyperparameter tuner for <code>QRFGPR</code> models.</td>
177+
</tr></tbody>
178+
</table></div></div>
179+
119180
</section>
120181

121182
<section class="container py-3">
@@ -125,16 +186,6 @@ <h2 class="py-2" id="rfflearn_gpu">rfflearn.gpu</h2>
125186
-->See the <a href="#rfflearn_cpu">API reference of rfflearn.cpu</a> for the details of this module.</p>
126187
</section>
127188

128-
<section class="container py-3">
129-
<h2 class="py-2" id="rfflearn_explainer">rfflearn.explainer</h2>
130-
<p>TBD</p>
131-
</section>
132-
133-
<section class="container py-3">
134-
<h2 class="py-2" id="rfflearn_tuner">rfflearn.tuner</h2>
135-
<p>TBD</p>
136-
</section>
137-
138189
</main>
139190

140191
<!-- Footer -->

0 commit comments

Comments
 (0)