Skip to content

[Katib] Update documentation for suggestion algorithms #3418

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 2 commits into from
Jan 5, 2023
Merged

[Katib] Update documentation for suggestion algorithms #3418

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
b87309c
Update details for the grid search algorithm
tenzen-y Dec 24, 2022
ae5293a
Remove documentations for chocolate suggestion service
tenzen-y Dec 24, 2022
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
28 changes: 23 additions & 5 deletions content/en/docs/components/katib/experiment.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -246,9 +246,29 @@ continuous) and the number of possibilities is low. A grid search
performs an exhaustive combinatorial search over all possibilities,
making the search process extremely long even for medium sized problems.

Katib uses the [Chocolate](https://chocolate.readthedocs.io) optimization
Katib uses the [Optuna](https://github.com/optuna/optuna) optimization
framework for its grid search.

<div class="table-responsive">
<table class="table table-bordered">
<thead class="thead-light">
<tr>
<th>Setting name</th>
<th>Description</th>
<th>Example</th>
</tr>
</thead>
<tbody>
<tr>
<td>random_state</td>
<td>[int]: Set <code>random_state</code> to something other than None
for reproducible results.</td>
<td>10</td>
</tr>
</tbody>
</table>
</div>

<a id="random-search"></a>

#### Random search
Expand All @@ -263,8 +283,7 @@ use when combinatorial exploration is not possible. If the number of continuous
variables is high, you should use quasi random sampling instead.

Katib uses the [Hyperopt](https://hyperopt.github.io/hyperopt/),
[Goptuna](https://github.com/c-bata/goptuna),
[Chocolate](https://chocolate.readthedocs.io) or
[Goptuna](https://github.com/c-bata/goptuna) or
[Optuna](https://github.com/optuna/optuna) optimization
framework for its random search.

Expand Down Expand Up @@ -306,8 +325,7 @@ steps, making it a good choice when the time to
complete the evaluation of a parameter configuration is long.

Katib uses the
[Scikit-Optimize](https://github.com/scikit-optimize/scikit-optimize) or
[Chocolate](https://chocolate.readthedocs.io) optimization framework
[Scikit-Optimize](https://github.com/scikit-optimize/scikit-optimize) optimization framework
for its Bayesian search. Scikit-Optimize is also known as `skopt`.

Katib supports the following algorithm settings:
Expand Down
7 changes: 1 addition & 6 deletions content/en/docs/components/katib/katib-config.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -182,11 +182,6 @@ any other settings, a default value is set automatically.
<td><code>random</code>, <code>tpe</code></td>
<td><a href="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/hyperopt/hyperopt">Hyperopt</a> optimization framework</td>
</tr>
<tr>
<td><code>suggestion-chocolate</code></td>
<td><code>grid</code>, <code>random</code>, <code>quasirandom</code>, <code>bayesianoptimization</code>, <code>mocmaes</code></td>
<td><a href="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/AIworx-Labs/chocolate">Chocolate</a> optimization framework</td>
</tr>
<tr>
<td><code>suggestion-skopt</code></td>
<td><code>bayesianoptimization</code></td>
Expand All @@ -199,7 +194,7 @@ any other settings, a default value is set automatically.
</tr>
<tr>
<td><code>suggestion-optuna</code></td>
<td><code>multivariate-tpe</code>, <code>tpe</code>, <code>cmaes</code>, <code>random</code></td>
<td><code>multivariate-tpe</code>, <code>tpe</code>, <code>cmaes</code>, <code>random</code>, <code>grid</code></td>
<td><a href="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/optuna/optuna">Optuna</a> optimization framework</td>
</tr>
<tr>
Expand Down