BTB
BTB copied to clipboard
MLBazaar
A simple, extensible library for developing AutoML systems
Create the following documentation pages: - getting started: intro - getting started: installation - getting started: rename basic usage -> quickstart, add a bit on hyperparameters - advanced usage: basic...
The changes involved are: - Change the `setup.py` python specifications - Update `copula` to the latest version - Add Python3.9 to `tox.ini` - Add Python3.9 to `.github/workflows/tests.yml` - Add 3.9...
* BTB version: 0.3.12 ### Description Examine distribution estimation of univariates when using copulas due to small sample size.
The ```bayesmark``` package is another ```wrapper``` hyper parameter tuning library. We can add this to our benchmarking suite. Per their documentation, they wrap around: ``` The builtin optimizers are wrappers...
In our collection of datasets that we use for our benchmarking process, we find some datasets that do not represent a hard problem and exhibit the following properties when solving...
At the moment we have three mathematical functions as `Challenges`: - Rosenbrock - Branin - Bohachevsky We would like to review them and make sure they are still good for...
Nowadays most of the `Hyperparameters` are using a multi type value where their value can take several values, at the moment `BTB` is missing a `MultiTypeHyperparam` class. These hyperparameters consist...
The [Matern 5/2 kernel](https://scikit-learn.org/stable/modules/generated/sklearn.gaussian_process.kernels.Matern.html#sklearn.gaussian_process.kernels.Matern) was proposed in [1] as a good choice of GP kernel due to several advantageous properties. I propose to implement a `GPMatern52EiTuner` within btb that uses...
* BTB version: 06daf139b88854c727d352a3c9dc919760e02fd6 * Python version: 3.7.2 * Operating System: Debian stretch ### DescriptionFile ``` "lib/python3.7/site-packages/btb/tuning/gp.py", line 86, in _acquire ei = sigma * (z * Phi(z) + N(z))...
