Hi Matthias.
As far as I know, the main goal for TPE was to support tree-structured
parameter spaces. I am not sure we want to go there yet because of the
more complex API.
On non-tree structured spaces, I think TPE performed worse than SMAC and GP.
With regard to your code: There might be touchy legal issues involved if
you didn't publish your code and we base our implementation on it.
If your code is public and BSD / MIT licensed, it would probably be much
safer. Why don't you just push your code under a permissive license?
Thank you for providing your benchmarks, they might be quite helpful.
Cheers,
Andy
On 03/26/2015 11:17 AM, Matthias Feurer wrote:
Dear Christof, dear scikit-learn team,
This is a great idea, I highly encourage your idea to integrate
Bayesian Optimization into scikit-learn since automatically
configuring scikit-learn is quite powerful. It was done by the three
winning teams of the first automated machine learning competition:
https://sites.google.com/a/chalearn.org/automl/
I am writing this e-mail because our research group on learning,
optimization and automated algorithm design
(http://aad.informatik.uni-freiburg.de/) is working on very similar
things which might be useful in this context. Some people in our lab
(together with some people from other universities)developed a
framework for robust Bayesian optimization with minimal external
dependencies. It currently depends on GPy, but this dependency could
be easily replaced by the scikit-learn GP. It is probably not as
leightweight as you want to have it for scikit-learn, but you might
want to have a look at the source code. I will provide a link as soon
as the project is public (which is soon). In the meantime, I can grant
read-access to those who are interested. It might be helpful for you
to have look at the structure of the module.
Besides these remarks, I think that using a GP is a good way to tune
the few hyperparameters of a single model. Another remark: Instead of
comparing GPSearchCV to spearmint only, you should also consider the
TPE algorithm implemented in hyperopt
(https://github.com/hyperopt/hyperopt). You could consider the
following benchmarks:
1. Together with a fellow student I implemented a library called
HPOlib, which provides a few benchmarks for hyperparameter
optimization (for example some from the 2012 spearmint paper):
https://github.com/automl/HPOlib It is further described in this
paper: http://automl.org/papers/13-BayesOpt_EmpiricalFoundation.pdf
2. If you are looking for a small pipeline, you can use
sklearn.feature_selection.SelectPercentile with a fixed scoring
function together with a classification algorithm. It adds a single
hyperparameter which should be a good fit for the GP.
Best regards,
Matthias
------------------------------------------------------------------------------
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the
conversation now. http://goparallel.sourceforge.net/
_______________________________________________
Scikit-learn-general mailing list
Scikit-learn-general@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
------------------------------------------------------------------------------
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the
conversation now. http://goparallel.sourceforge.net/
_______________________________________________
Scikit-learn-general mailing list
Scikit-learn-general@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
