search

SheffieldML / GPy

Gaussian processes framework in python
BSD 3-Clause "New" or "Revised" License
2.05k stars 562 forks source link

Moved, delete this #755

Open KOLANICH opened 5 years ago

KOLANICH commented 5 years ago

Moved to https://github.com/SheffieldML/GPyOpt/issues/237

mzwiessele commented 5 years ago

It’s the other dimension which needs to be one: Y.shape == (n, 1)

X can be multidimensional: X.shape == (n, q)

On 15 May 2019, at 07:16, KOLANICH notifications@github.com wrote:

My code

f is set later

optimizer = GPyOpt.methods.BayesianOptimization(f=None, domain=specSeq, ... X=np.empty((0, len(specSeq))), Y=np.empty((0, 1)), ...) results in

File "%SECRET%\lib\site-packages\GPyOpt\core\bo.py", line 137, in run_optimization self._update_model(self.normalization_type) File "%SECRET%\lib\site-packages\GPyOpt\core\bo.py", line 249, in _update_model Y_inmodel = normalize(self.Y, normalization_type) File "%SECRET%\lib\site-packages\GPyOpt\util\general.py", line 215, in normalize raise NotImplementedError('Only 1-dimensional arrays are supported.') NotImplementedError: Only 1-dimensional arrays are supported. If I change to


optimizer = GPyOpt.methods.BayesianOptimization(f=None, domain=specSeq, ... X=np.empty((0, len(specSeq))), Y=np.empty(0), ...) 
then I get

  File "%SECRET%\lib\site-packages\GPy\core\gp.py", line 48, in __init__
    assert Y.ndim == 2
—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub, or mute the thread.
wleoncio commented 3 years ago

Looks like this issue should be closed (maybe also tagged as "duplicated")