search

josejimenezluna / pyGPGO

Bayesian optimization for Python
http://pygpgo.readthedocs.io
MIT License
241 stars 62 forks source link

theano and cov.py warnings #33

Open Daniel-Trung-Nguyen opened 3 years ago

Daniel-Trung-Nguyen commented 3 years ago

I ran into these warnings when running the tutorials on Google Colab. Do these signal version incompatibility for some libraries?

_Evaluation Proposed point Current eval. Best eval. /usr/local/lib/python3.7/dist-packages/pymc3/gp/cov.py:97: UserWarning: Only 1 column(s) out of 2 are being used to compute the covariance function. If this is not intended, increase 'input_dim' parameter to the number of columns to use. Ignore otherwise. UserWarning, WARNING (theano.tensor.blas): We did not find a dynamic library in the library_dir of the library we use for blas. If you use ATLAS, make sure to compile it with dynamics library. Only 300 samples in chain. Sequential sampling (2 chains in 1 job) CompoundStep

Slice: [log_s2_n] Slice: [log_s2_f] Slice: [l]

100.00% [1300/1300 00:05<00:00 Sampling chain 0, 0 divergences]

100.00% [1300/1300 00:03<00:00 Sampling chain 1, 0 divergences] Sampling 2 chains for 1_000 tune and 300 draw iterations (2_000 + 600 draws total) took 10 seconds. The estimated number of effective samples is smaller than 200 for some parameters. init [0.5881308 0.89771373]. 0.12055515791243088 0.35562010326493304 init [0.89153073 0.81583748]. 0.06525608955367107 0.35562010326493304 init [0.03588959 0.69175758]. 0.35562010326493304 0.35562010326493304 /usr/local/lib/python3.7/dist-packages/pymc3/gp/cov.py:97: UserWarning: Only 1 column(s) out of 2 are being used to compute the covariance function. If this is not intended, increase 'input_dim' parameter to the number of columns to use. Ignore otherwise. UserWarning, Only 300 samples in chain. Sequential sampling (2 chains in 1 job) CompoundStep

Slice: [log_s2_n] Slice: [log_s2_f] Slice: [l]

100.00% [1300/1300 00:04<00:00 Sampling chain 0, 0 divergences]

100.00% [1300/1300 00:04<00:00 Sampling chain 1, 0 divergences] Sampling 2 chains for 1_000 tune and 300 draw iterations (2_000 + 600 draws total) took 8 seconds. 1 [0. 0.]. 0.7664205912849231 0.7664205912849231 /usr/local/lib/python3.7/dist-packages/pymc3/gp/cov.py:97: UserWarning: Only 1 column(s) out of 2 are being used to compute the covariance function. If this is not intended, increase 'inputdim' parameter to the number of columns to use. Ignore otherwise. UserWarning,

Daniel-Trung-Nguyen commented 3 years ago

Also, when running the gpgo.GP.posteriorPlot() block, I got this error:

TypeError Traceback (most recent call last)

in () ----> 1 gpgo.GP.posteriorPlot() 1 frames /usr/local/lib/python3.7/dist-packages/pyGPGO/surrogates/GaussianProcessMCMC.py in posteriorPlot(self) 84 """ 85 with self.model as model: ---> 86 pm.traceplot(self.trace, varnames=['l', 'sigmaf', 'sigman']) 87 plt.tight_layout() 88 plt.show() /usr/local/lib/python3.7/dist-packages/pymc3/plots/__init__.py in wrapped(*args, **kwargs) 37 if "varnames" in kwargs: 38 raise DeprecationWarning( ---> 39 f"The `varnames` kwarg was renamed to `var_names`.", stacklevel=2 40 ) 41 original = func.__name__ TypeError: DeprecationWarning() takes no keyword arguments
josejimenezluna commented 3 years ago

Dear @Daniel-Trung-Nguyen,

This probably means that BLAS is not installed on the system. How did you install the package? did you have a previous version of numpy installed?

josejimenezluna commented 3 years ago

The varnames issue should be fixed with https://github.com/josejimenezluna/pyGPGO/commit/587d03a20a60bc135c913849fcfd01b1354ef6e8

Daniel-Trung-Nguyen commented 3 years ago

@josejimenezluna, I just got it fired up on Google Colab which comes with preinstalled packages on a Linux system. One unrelated question, but can you explain the I, sigmaf, sigman ouput variables on the gpgo.GP.posteriorPlot() function? Thanks, Daniel