search

SheffieldML / GPy

Gaussian processes framework in python
BSD 3-Clause "New" or "Revised" License
2.04k stars 562 forks source link

2D Regression start point #884

Open riazcseiu opened 3 years ago

riazcseiu commented 3 years ago

I have a dataset and corresponding data flag for a Wafer. The structure is as follows like :

Data_set sample 

[[NaN NaN NaN NaN .....3456 .8765 .56454 .....NaN NaN ]
     [NaN NaN NaN .....765 .3456 .8765 .56454 ..... NaN NaN]
-----
     [NaN NaN NaN NaN... .3456 .8765 .56454 -----NaN NaN]]

and the Data_flag is as follows : 

   [[0 0 0 0 ....1 1 1.... 0 0]
     [0 0 0 ....1 1 1 1 ....0 0]
----
     [0 0 0 0 ...1 1 1.. 0 0]]

So the in the DataFlag the NaN is replaced by 0 and the other position is replaced by 1. I want to select a few positions from the Data_set based on the value of the Data_flag (where the value is 1) and then I want to predict the other position value based on data_flag. I want to solve using this framework.

I'm confused about which solution is applicable for me form https://gpy.readthedocs.io/en/deploy/_modules/GPy/examples/regression.html#multioutput_gp_with_derivative_observations

I also tried [here] https://github.com/stocon/pycon2016

I'm rocky and do a lot of search for the perfect start point of such a problem. please suggest start point Thank you

ekalosak commented 3 years ago

Sorry, your problem isn't entirely clear. Would you be able to post a code snippet and some sample data so we can be more helpful?